{ "cells": [ { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "70259e95e86691d4999fbd5761a7f211", "grade": false, "grade_id": "cell-8a7cd222d4950599", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "## Exercise Sheet No. 3\n", "\n", "---\n", "\n", "> Machine Learning for Natural Sciences, Summer 2024, T.T.-Prof. Pascal Friederich, pascal.friederich@kit.edu\n", "> \n", "> Deadline: Monday 06.05.2024, 8:00 am\n", "> \n", "> Tutor: luca.torresi@kit.edu \n", "> **Please ask questions in the forum and only contact the Tutor when there are issues with the grading**\n", "---\n", "\n", "**Topic**: This exercise sheet will focus on linear algebra, knn classifier, precision, recall, ROC curves and feature reduction. You will continue to use ``numpy`` methods." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Please add here your group members' names and student IDs (up to 3 members per team). \n", "\n", "Names: Nils Lennart Bruns\n", "\n", "IDs: usxfs\n", "\n", "**Each student has to submit their own notebook, even when they worked in a team**" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "f8ffb30bf172b0c558a7d62470462174", "grade": false, "grade_id": "cell-c26ffc5de714a368", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "## 3.1 Distance function computation" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "15c883f32c8106c2c326e4953f5d5f81", "grade": false, "grade_id": "cell-4bf94758d89ed7fa", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "It often happens, whether in the implementation of ML models or when calculating feature descriptors, that we need to compute the mutual Euclidean distance between a set of points. The Euclidean distance of two given data points can serve as a measure of how differently we think they will behave. In this exercise we will focus on an efficient `numpy` implementation of the mutual Euclidean distance between a set of points.\n", "\n", "**3.1.1** Implement a function ``dist_loop(A,B)`` to compute the Euclidean distance between all elements of two sets of points $A \\subset \\mathbb{R}^{D}$ and $B \\subset \\mathbb{R}^{D}$. Use explicit python loops to find the distance $d_{ij} = || a_{i} - b_{j} ||$ between all points $a_{i} \\in A$ and $b_{j} \\in B$. The input should be two matrices of shape $N \\times D$ and $M \\times D$. The output should be a $N \\times M$ distance matrix. For the calculation of the Euclidean distance you might want to use ``numpy.square()``, ``numpy.sum()`` and ``numpy.sqrt()``." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "90e6ddbc23e3e3b9b41fbb33d6f9be05", "grade": false, "grade_id": "cell-c33edbe7e3ce1318", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "A shape: (500, 3) B shape: (1000, 3)\n" ] } ], "source": [ "##### DO NOT CHANGE #####\n", "\n", "import numpy as np\n", "from numpy import random\n", "import sys\n", "n = 500\n", "m = 1000\n", "d = 3\n", "A_data = np.reshape(random.rand(n*d),(n,d))\n", "B_data = np.reshape(random.rand(m*d),(m,d))\n", "print(\"A shape:\", A_data.shape, \"B shape:\", B_data.shape)\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "8501f3bf68a844bf562dc6c0f94bb1d3", "grade": false, "grade_id": "dist_loop", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [ { "data": { "text/plain": [ "array([[0.80885463, 0.8195383 , 0.71226228, ..., 0.61804707, 0.8016523 ,\n", " 1.13608461],\n", " [0.67290947, 1.08141804, 1.12906026, ..., 0.70877396, 0.56271374,\n", " 1.06855972],\n", " [0.25002696, 0.76889124, 0.92001653, ..., 0.93756795, 0.65637306,\n", " 0.32381401],\n", " ...,\n", " [0.63405928, 1.06695014, 0.99081191, ..., 1.09981833, 0.0916953 ,\n", " 0.85433308],\n", " [0.86210548, 0.99582835, 0.9468733 , ..., 0.61701206, 0.79183323,\n", " 1.23683335],\n", " [0.41960192, 0.79818752, 0.75174524, ..., 0.89800484, 0.31068329,\n", " 0.66044397]])" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def dist_loop(A,B):\n", " if A.shape[-1] != B.shape[-1]:\n", " raise ValueError(\"Error A and B must have same last dimension but got\", A.shape[-1], B.shape[-1])\n", "\n", " matrix = []\n", " for a in A:\n", " v = []\n", " for b in B:\n", " s = 0\n", " for i in range(A.shape[-1]):\n", " s += np.square(a[i]-b[i])\n", " v.append(np.sqrt(s))\n", " matrix.append(v)\n", " return np.array(matrix)\n", " \n", "\n", "dist_loop(A_data, B_data)" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "8dfd741f6e837714cccd71879127ded0", "grade": false, "grade_id": "cell-fa9a5e0893f92ee1", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "**3.1.2** Loops are rather slow in Python, so we would like to have more time-efficient code for our numeric operations. Let's write another function ``dist_vec(A,B)`` to compute the distance relying on vectorization with ``numpy`` methods. Consult https://www.safaribooksonline.com/library/view/python-for-data/9781449323592/ch04.html and https://softwareengineering.stackexchange.com/questions/254475/how-do-i-move-away-from-the-for-loop-school-of-thought if you need more information on how to do this. \n", "\n", "Tip: There is more than one solution. One possible solution involves adding additional dimensions to `np.arrays`, for example using `expand_dims`. In fact, when two `np.arrays` do not have matching shapes they are automatically broadcasted by repeating their respective element along the axis in question. Beside broadcasting, also the [indexing rules](https://numpy.org/doc/stable/reference/arrays.indexing.html) (required for 3.2, 3.3) of `numpy` help with vectorization to avoid python loops. For this task, you cannot use any ``scipy`` method!" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "c528a8f299889fc60698e8e2ca273aad", "grade": false, "grade_id": "dist_vec", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [], "source": [ "def dist_vec(A,B):\n", " #print(A.shape, B.shape)\n", " if A.shape[-1] != B.shape[-1]:\n", " raise ValueError(\"Error A and B must have same last dimension but got\", A.shape[-1], B.shape[-1])\n", " B=np.tile(B.reshape((1,)+B.shape),(A.shape[0],1,1))\n", " A=np.tile(A.reshape((A.shape[0],1,A.shape[1])),(1,B.shape[1],1))\n", " #print(B.shape, A.shape)\n", " res = np.sqrt(np.sum(np.square(B - A), axis=2))\n", " #print(res.shape)\n", " return res\n", "\n", "#%timeit -n 10 result_vec = dist_vec(A_data,B_data)\n", "\n", "\"\"\"\n", "def dist_vec(A,B):\n", " #print(A.shape, B.shape)\n", " #A = A.reshape((500, 1,3))\n", " A = np.expand_dims(A, axis=1)\n", " #B = B.reshape((1, 1000,3))\n", " B = np.expand_dims(B, axis=0)\n", " #print(A.shape, B.shape)\n", " A-B\n", " res = np.sqrt(np.sum(np.square(B - A), axis=2))\n", " return res\n", "\"\"\"\n", "\n", "#%timeit -n 10 result_vec = dist_vec(A_data,B_data)\n", "None" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "4824f5c64a385e4cf3b0a76cf490a9ef", "grade": false, "grade_id": "cell-fc0ec9c589763c8a", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Let's check that the two functions return the same result. Then for each method, assign the shape of the output to the respective variable." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "d4b043d946187d5d4b08030dc999dd25", "grade": false, "grade_id": "difference_dist", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [], "source": [ "result_loop = dist_loop(A_data,B_data)\n", "result_vec = dist_vec(A_data,B_data)\n", "result_vec_shape = None # Check shape of result_loop\n", "result_loop_shape = None # Check shape of result_vec\n", "\n", "result_vec_shape = result_vec.shape # Check shape of result_loop\n", "result_loop_shape = result_loop.shape # Check shape of result_vec\n", "assert result_vec_shape == result_loop_shape\n", "assert (result_loop == result_vec).all()" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "e12995c578ef850b181a8480d7efb5e4", "grade": false, "grade_id": "cell-b4c698e9d03d3613", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Now compare the run times of the two implementations using jupyter's ``%timeit`` command or pythons ``time``. How much faster is the vectorized version?" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "a4b2b3eb3f759651f178003bf2c921c2", "grade": false, "grade_id": "cell-86d6de253ac2af1b", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "11.8 s ± 332 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n", "62.3 ms ± 3.57 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)\n" ] } ], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# measure time for loop\n", "%timeit result_loop = dist_loop(A_data,B_data)\n", "# measure time for vectorized\n", "%timeit result_vec = dist_vec(A_data,B_data)\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "eac19d845e37926fbae4480a65036adf", "grade": false, "grade_id": "speed_up", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "232.2784810126582\n" ] } ], "source": [ "speed_up_factor = None # A rough estimation is okay.\n", "\n", "speed_up_factor = 3.67 / 0.0158\n", "print(speed_up_factor)" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "efc96f4754e60fae045f165cf37d914e", "grade": false, "grade_id": "cell-a20dc8356bb8b477", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Note: It is very important that you understand vectorization, indexing and broadcasting. All deep learning frameworks are based on tensor (array) operations, just like the one you implemented using ``numpy``." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "f52d547e2b5c93e53ac3b04177e56fea", "grade": true, "grade_id": "test_funs_1", "locked": true, "points": 3, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# Test for grading\n", "assert callable(dist_loop)\n", "# Hidden tests check with random inputs that the function returns the correct values\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "118c673fb80f132ee673c6a7560e3e3b", "grade": true, "grade_id": "test_funs_2", "locked": true, "points": 3, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "assert callable(dist_vec)\n", "# Hidden tests check with random inputs that the function returns the correct values\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "84c003a9b466a40c1432105e23773ce8", "grade": true, "grade_id": "test_loop_vec_diff", "locked": true, "points": 1, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# Test for grading\n", "assert result_loop is not None\n", "assert result_vec is not None\n", "# Hidden tests to check that the functions returned same values\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "cee3584343bd5749f71cba05f1b14aa0", "grade": true, "grade_id": "test_shape_dist", "locked": true, "points": 2, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "assert result_vec_shape is not None\n", "assert result_loop_shape is not None\n", "# Hidden tests to check that the functions returned outputs with correct shape\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "270b8a913d4524118f5b90feb704e9c9", "grade": true, "grade_id": "speed_up_test", "locked": true, "points": 1, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# Test for grading\n", "assert speed_up_factor > 0\n", "# Hidden test to check if speed up factor lies on an expected interval\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "fc2738648d5a0815497e45f7ae8687f8", "grade": false, "grade_id": "cell-cae27f50792e197e", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "## 3.2 Precision-Recall Curves" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "f4ef8219eba10558f4ab839dfda682a9", "grade": false, "grade_id": "cell-9f173d428aa9d6b2", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "In this section, we will use sklearn's digits dataset (64 pixels, grey scale) for an image classification task.
\n", "We will use a k-nearest neighbors (k-NN) classifier to discriminate the images depending on the digit they represent. We will keep a small subset of the dataset to train our classifier and then use our trained model on the remaining images.
\n", "To apply k-NN we need to choose a measure of similarity: we will use the previously implemented function to compute the Euclidian distance between couples of image vectors.
\n", "Let's start loading the dataset, splitting it in a train and a test set and having a look at some of the images." ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "13cae695b390766e072d1eba05177fcb", "grade": false, "grade_id": "cell-4694919154d6a077", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train set shape: (35, 64); test set shape: (1762, 64)\n", "\n", "\n", "Example of images from classes: [0 1 2]\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "##### DO NOT CHANGE #####\n", "\n", "import numpy as np\n", "import pandas as pd\n", "from sklearn.datasets import load_digits\n", "import matplotlib.pyplot as plt\n", "from sklearn.model_selection import train_test_split\n", "\n", "# read and prepare the digits data\n", "digits = load_digits()\n", "data = digits[\"data\"]\n", "images = digits[\"images\"]\n", "target = digits[\"target\"]\n", "\n", "# split data in train and test set\n", "X_labeled, X_unknown, y_labeled, y_unknown = train_test_split(data, target, test_size=0.98, random_state=42)\n", "print(f'Train set shape: {X_labeled.shape}; test set shape: {X_unknown.shape}\\n\\n')\n", "\n", "# Show digits\n", "print(f'Example of images from classes: {target[:3]}')\n", "fig = plt.figure(figsize = (10,3))\n", "plt.gray()\n", "plt.subplot(1,3,1); plt.axis('off')\n", "plt.imshow(images[0])\n", "plt.subplot(1,3,2); plt.axis('off')\n", "plt.imshow(images[1])\n", "plt.subplot(1,3,3); plt.axis('off')\n", "plt.imshow(images[2])\n", "fig.tight_layout(); plt.show()\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "cb87fc8d9e0cbaa69850ae51b86ff3c3", "grade": false, "grade_id": "cell-3a005da3224b0a6c", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "$\\textbf{k-NN}$ is a simple non-parametric model. For each datapoint in the test set we want to build a list, sorted by the chosen similarity measure, of all the datapoints in the train set. Then the output of the classifier is determined by the \"votes\" of the k closest train datapoints, of which we assume we know the correct label. For example we can simply select as predicted class the most common class among the closest neighbors, with a majority vote. Or, in a binary classification setting, define the probability of an object to belong to a given class as the normalized number of positive votes and then predict positively if the probability is over a pre-set threshold.\n", "\n", "Let's implement a simple k-NN classifier that can work both for a multiclass setting and as binary classifier when we specify a digit of interest. We will use the binary classifier to discuss some common performance metrics later on.\n", "\n", "As introduced above, we will use the Euclidean distance between pixel values as similarity measure:\n", "\n", "$d(X_i,X_{i'} ) = || X_i - X_{i'}||_2$\n", "\n", "To efficiently compute these distances, you should use vectorization from exercise 3.1. Let $D$\n", "be the full dissimilarity matrix, i.e. $D_{i i'} = d(X_i,X_{i'})$. An ``np.argsort()`` of row $D_i$ now gives the similarity ordering, relative to image $X_i$, of all images $X_{i'}$ in the train set.\n", "\n", "\n", "Note: If you didn't manage to solve 3.1 you can now use `scipy.spatial.distance.cdist`. But you do not get points for this solution in 3.1." ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "06fdd033b8bbf6d9b51a851cc4bd69e8", "grade": false, "grade_id": "knn_classifier", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [], "source": [ "from collections import Counter\n", "from scipy.spatial.distance import cdist\n", "\n", "\n", "class Knn_Classifier:\n", " def __init__(self, X_unknown, X_labeled, y_labeled):\n", " D = None # compute distances from each X_unknown to each X_labeled \n", " nn = None # order neighbors from closest to furthest\n", " self.nn_targets = None # store labels (our 'votes') of the ordered neighbors (given by y_labeled)\n", "\n", " D = dist_vec(X_unknown, X_labeled)\n", " nn = np.argsort(D)\n", " self.nn_targets = y_labeled[nn]\n", "\n", " def set_k(self, k):\n", " self.knn_targets = None # restrict list of labels to k closest neighbors\n", " self.knn_targets = self.nn_targets[:,:k]\n", " \n", " def __call__(self, label: int = None, threshold: float = .5):\n", " counters = [Counter(self.knn_targets[i]) for i in range(self.knn_targets.shape[0])]\n", " if label is None:\n", " # multiclass classifier, for each data point it outputs the most \"voted\" class by the neighbors\n", " preds = np.asarray([c.most_common(1)[0][0] for c in counters])\n", " return preds\n", " else:\n", " # binary classifier for digit 'label'\n", " # computes the probability a given data point belongs to a class or to any of the others\n", " # as the number of votes for that class normalized over the total number of votes\n", " probs = np.asarray([c.get(label, 0)/sum(c.values()) for c in counters])\n", " # predicts as belonging to the class 'label' if the computed probability exceeds the threshold\n", " preds = (probs>=threshold).astype(int)\n", " return preds, probs" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "5ea00eff0afabd84b9b1e8cdf07b52ba", "grade": false, "grade_id": "cell-0c1571f58a01fa3b", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Let's run a multiclass knn classification setting the number of neighbors k to 5. \n", "\n", "To have an overview of the goodness of the classification we will print a $\\textbf{confusion matrix}$. A confusion matrix is defined as the matrix $C$ such that $C_{ij}$ is equal to the number of observations known to be in group $i$ and predicted to be in group $j$. It can be easily computed using the function \"confusion_matrix\" from sklearn.metrics and we'll plot it using seaborn heatmaps." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "7d3f3d749569eb954ca238137676633e", "grade": false, "grade_id": "cell-1b7c0cef909cd1ca", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "##### DO NOT CHANGE #####\n", "\n", "import seaborn as sns\n", "from sklearn.metrics import confusion_matrix\n", "\n", "knn = Knn_Classifier(X_unknown, X_labeled, y_labeled)\n", "knn.set_k(5)\n", "preds = knn()\n", "cf_matrix = confusion_matrix(y_unknown, preds)\n", "\n", "fig, ax = plt.subplots(figsize=(8,6))\n", "ax = sns.heatmap(cf_matrix, annot=True, fmt='', cmap='Blues')\n", "plt.xlabel('Predicted labels')\n", "plt.ylabel('Ground truth labels')\n", "plt.show()\n", "plt.close()\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "5cbe40e883814d999a5e61922e1d158d", "grade": false, "grade_id": "cell-e875203b21c8141c", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Each row of the confusion matrix represents the instances in an actual class while each column represents the instances in a predicted class. Once produced, we can use it directly to compute some performance measures. We will have a look at two evaluation metrics used to assess the performance of a machine learning model in a binary classification problem: precision and recall. \n", "\n", "In a binary setting, so considering in our example one class against all other classes, we can define:\n", "
    \n", "
  1. true positives (TP): number of objects belonging to the class that were $\\textbf{correctly}$ predicted as belonging to it;

    \n", "
  2. true negatives (TN): number of objects $\\textbf{not}$ belonging to the class that were $\\textbf{correctly}$ predicted as $\\textbf{not}$ belonging to it;

    \n", "
  3. false positives (FP): number of objects $\\textbf{not}$ belonging to the class that were $\\textbf{incorrectly}$ predicted as belonging to it;

    \n", "
  4. false negatives (FN): number of objects belonging to the class that were $\\textbf{incorrectly}$ predicted as $\\textbf{not}$ belonging to it;

    \n", "
\n", "\n", "Given these four quantities we can define:\n", "
    \n", "
  1. $\\textbf{Precision}$, or positive predictive value, is the proportion of true positive predictions among all the positive predictions made by the model. In formulas: $\\text{precision} = \\frac{\\text{TP}}{\\text{TP} + \\text{FP}}$

    \n", "\n", "
  2. $\\textbf{Recall}$, also termed sensitivity or probability of detection, is the proportion of true positive predictions among all the actual positive cases in the dataset. In formulas: $\\text{recall} = \\frac{\\text{TP}}{\\text{TP} + \\text{FN}}$.

    \n", "
\n", "\n", "\n", "A high precision indicates that the model is making few false positive predictions, i.e., positive predictions are correct. A high recall indicates that the model is making few false negative predictions, i.e., the model is correctly identifying all positive cases in the dataset. Ideally we would like both of these measures to be equal to 1, meaning that all objects predicted as belonging to the class are actually from that class (precision=1) and all objects belonging to the class are correctly predicted as belonging to it (recall=1). In general, there is a trade-off between precision and recall: increasing one metric often comes at the expense of the other. \n", "\n", "Let's compute precision and recall of our multiclass k-NN classifier using the confusion matrix we computed above." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performance of multiclass classifier for digit 0 VS all others:\n", "\t-True positives: 174\n", "\t-False positives: 27\n", "\t-False negatives: 1\n", "\t-Precision: 0.87\n", "\t-Recall: 0.99\n", "Performance of multiclass classifier for digit 1 VS all others:\n", "\t-True positives: 135\n", "\t-False positives: 59\n", "\t-False negatives: 44\n", "\t-Precision: 0.70\n", "\t-Recall: 0.75\n", "Performance of multiclass classifier for digit 2 VS all others:\n", "\t-True positives: 107\n", "\t-False positives: 27\n", "\t-False negatives: 66\n", "\t-Precision: 0.80\n", "\t-Recall: 0.62\n" ] } ], "source": [ "for label in range(3):\n", " print(f'Performance of multiclass classifier for digit {label} VS all others:')\n", " # to access predictions relative to a given class i we select the i-th column of the confusion matrix\n", " column_counts = cf_matrix[:, label]\n", " # the i-th row corresponds to the images actually belonging to class i \n", " row_counts = cf_matrix[label]\n", "\n", " TP = column_counts[label]\n", " FP = sum(column_counts) - column_counts[label]\n", " FN = sum(row_counts) - row_counts[label]\n", " \n", " precision = TP/(TP+FP)\n", " recall = TP/(TP+FN)\n", "\n", " print(f'\\t-True positives: {TP}')\n", " print(f'\\t-False positives: {FP}')\n", " print(f'\\t-False negatives: {FN}')\n", " print(f'\\t-Precision: {precision:.2f}')\n", " print(f'\\t-Recall: {recall:.2f}')" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "6abefaab24a40a0a26c906e6fd9d9ead", "grade": false, "grade_id": "cell-9706dc8a04420cff", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "We have seen how to compute precision and recall with respect to each digit using the confusion matrix produced in our multiclass setting. We are now going to switch to binary classification and metrics to introduce a useful tool, the receiver operating characteristic curve, or ROC curve. The ROC curve is a graphical representation of the performance of a binary classifier system as its discrimination threshold is varied. It plots the true positive rate (TPR), which is another name for the recall, against the false positive rate (FPR) at various classification thresholds.\n", "\n", "We need now to define a new measure. The FPR, also termed fall-out or false alarm ratio, measures the proportion of negative cases that are incorrectly classified as positive, in formulas $\\text{recall} = \\frac{\\text{FP}}{\\text{FP} + \\text{TN}}$.\n", "\n", "Let's implement a function that given a k-NN model, the label of interest and a classification threshold, computes all the performance measures we discussed until now." ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "44882e63e6953004576ac4e255771df8", "grade": false, "grade_id": "binary_metrics", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [], "source": [ "def compute_binary_metrics(knn, label: int, ground_truth: np.ndarray, threshold: float = .3):\n", " #print(list(ground_truth))\n", " metrics = {}\n", " preds, probs = knn(label, threshold)\n", " #print(list(preds), list(probs))\n", " target = (ground_truth==label).astype(int)\n", " metrics['TP'] = None # true_positives\n", " metrics['FP'] = None # false_positives\n", " metrics['TN'] = None # true_negative\n", " metrics['FN'] = None # false_negatives\n", "\n", " metrics['TP'] = sum(preds*target)\n", " metrics['FP'] = sum(preds*np.logical_not(target))\n", " metrics['TN'] = sum(np.logical_not(preds)*np.logical_not(target))\n", " metrics['FN'] = sum(np.logical_not(preds)*target)\n", " \n", " metrics['precision'] = metrics['TP']/(metrics['TP']+metrics['FP']) if metrics['TP']>0 else 0\n", " metrics['recall'] = metrics['TP']/(metrics['TP']+metrics['FN']) if metrics['TP']>0 else 0\n", " metrics['fall-out'] = None\n", "\n", " metrics['fall-out'] = metrics['FP']/(metrics['FP']+metrics['TN']) if metrics['FP']>0 else 0\n", "\n", " return metrics" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "8c6c3663ab02285ced39528e9a9dca25", "grade": false, "grade_id": "cell-ef9be3f98df88400", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Let's test our function and compare the metrics with the previously computed ones. \n", "\n", "What is different now, with respect to the multiclass setting, is that we can set a classification threshold. Check how changing the threshold changes the values of our metrics." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performance of binary classifier for digit 0:\n", "\t-TP: 174.00\n", "\t-FP: 1.00\n", "\t-TN: 1515.00\n", "\t-FN: 72.00\n", "\t-precision: 0.99\n", "\t-recall: 0.71\n", "\t-fall-out: 0.00\n", "Performance of binary classifier for digit 1:\n", "\t-TP: 140.00\n", "\t-FP: 39.00\n", "\t-TN: 1486.00\n", "\t-FN: 97.00\n", "\t-precision: 0.78\n", "\t-recall: 0.59\n", "\t-fall-out: 0.03\n", "Performance of binary classifier for digit 2:\n", "\t-TP: 147.00\n", "\t-FP: 26.00\n", "\t-TN: 1527.00\n", "\t-FN: 62.00\n", "\t-precision: 0.85\n", "\t-recall: 0.70\n", "\t-fall-out: 0.02\n" ] } ], "source": [ "# modify the threshold_ parameter within the interval (0,1)\n", "# see how the metrics are affected\n", "threshold_ = .3 \n", "knn.set_k(5)\n", "for label in range(3):\n", " print(f'Performance of binary classifier for digit {label}:')\n", " metrics = compute_binary_metrics(knn, label, y_unknown, threshold=threshold_)\n", " for k, v in metrics.items():\n", " print(f'\\t-{k}: {v:.2f}')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now, we can use these binary metrics to compute overall multi-class metrics using the following aggregation methods:
    \n", "
  1. Micro-averaging: In micro-averaging, we aggregate the binary metrics over all the classes. This means that we sum up the number of true positives, false positives, false negatives, and true negatives over all the classes and then compute the performance metrics.

    \n", "
  2. Macro-averaging: In macro-averaging, we compute the binary metrics, e.g. precision, for each class separately and then take the average across all the classes. Macro-averaging gives equal weight to all classes and is suitable when we want to evaluate the performance of each class separately.

    \n", "
  3. Weighted averaging: In weighted averaging, we compute the binary metrics for each class separately and then take the weighted average across all the classes. The weight of each class is proportional to the number of samples from that class in the dataset. Weighted averaging is most suitable averaging technique when the dataset is unbalanced among the various classes.

    \n", "
\n", "\n", "\n", "We can use the function below to compute multi-class performance measures, aggregating the corresponding binary metrics with weighted averaging. " ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "e19e1aa94b31762ba50c300b1066f333", "grade": false, "grade_id": "cell-d95e6fcea9692d4d", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "def compute_multiclass_metrics(knn, ground_truth, threshold=.3, labels=np.arange(10)):\n", " metrics_list = []\n", " for label in labels:\n", " metrics_list.append(compute_binary_metrics(knn, label, ground_truth, threshold))\n", " weights = np.asarray([(ground_truth==label).sum() for label in labels])\n", " multiclass_metrics = {}\n", " for m in ['precision', 'recall', 'fall-out']:\n", " multiclass_metrics[m] = sum([weights[i]*metrics_list[i][m] for i in range(len(labels))])/weights.sum()\n", " return multiclass_metrics, metrics_list\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Multiclass knn performance, weighted average: \n", "\t-precision: 0.84\n", "\t-recall: 0.64\n", "\t-fall-out: 0.02\n" ] } ], "source": [ "knn.set_k(5)\n", "metrics, _ = compute_multiclass_metrics(knn, y_unknown, threshold=.3)\n", "print(f'Multiclass knn performance, weighted average: ')\n", "for k, v in metrics.items():\n", " print(f'\\t-{k}: {v:.2f}')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As stated above, the ROC curve is generated by calculating the TPR and FPR for different threshold values, ranging from 0 to 1. As the threshold increases, the model becomes more conservative, classifying fewer instances as positive, which leads to a decrease in the FPR and TPR. Conversely, as the threshold decreases, the model becomes more aggressive, classifying more instances as positive, which leads to an increase in the FPR and TPR.\n", "\n", "A perfect classifier would have a ROC curve that passes through the top-left corner of the plot (TPR=1, FPR=0), indicating a TPR of 100% and an FPR of 0%. The area under the ROC curve (AUC) is a measure of the overall performance of the classifier, with a value of 1 indicating a perfect classifier and a value of 0.5 indicating a classifier that is no better than random guessing.\n", "\n", "The following code plots the Recall VS Precision (RvP) curve and the ROC curve (with AUC), so that we can see how modifying the threshold gives us classifiers with different trade-offs for the metrics." ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "03e3f6d3713e1ec9a7b7ecff034d6ab1", "grade": false, "grade_id": "cell-20702f8b17c1bd95", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "def plot_measures(knn, y_unknown, plot_type='ROC', all=True):\n", " measures = {'ROC':('fall-out', 'recall'), 'RvP': ('recall', 'precision')}\n", " measures = measures[plot_type]\n", " m0 = {k:[] for k in ['mc']+list(range(10))}\n", " m1 = {k:[] for k in ['mc']+list(range(10))}\n", " for threshold in np.arange(0,1,.1):\n", " mc, bin = compute_multiclass_metrics(\n", " knn, \n", " y_unknown, \n", " threshold)\n", " m0['mc'].append(mc[measures[0]])\n", " m1['mc'].append(mc[measures[1]])\n", " for i in range(10):\n", " m0[i].append(bin[i][measures[0]])\n", " m1[i].append(bin[i][measures[1]])\n", " if all:\n", " for i in range(10):\n", " plt.plot(m0[i][::-1], m1[i][::-1], label='digit_'+str(i))\n", " plt.plot(m0['mc'][::-1], m1['mc'][::-1], label='multiclass')\n", " auc = ''\n", " if plot_type=='ROC':\n", " plt.plot([0, 1], [0, 1], linestyle='dashed', color='red', label='random_guess')\n", " auc = np.abs(np.trapz(y=m1['mc'][::-1], x=m0['mc'][::-1]))\n", " auc = f', AUC: {auc:.3f}'\n", " plt.xlim(0, 1)\n", " plt.ylim(0, 1)\n", " plt.xlabel(measures[0])\n", " plt.ylabel(measures[1])\n", " plt.title(plot_type+auc)\n", " plt.legend()\n", " plt.grid()\n", " plt.show()\n", " plt.close()\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "8c7d14c3da4a8ef2a65620bd6c08a34d", "grade": false, "grade_id": "cell-ba8f5378c673ceef", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Let's plot some of the curves varying the number of neighbors we use for the classification." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "K: 2\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "K: 5\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "K: 20\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "for k in [2, 5, 20]:\n", " print(f'K: {k}')\n", " knn.set_k(k)\n", " plot_measures(knn, y_unknown, all=False, plot_type='RvP')\n", " plot_measures(knn, y_unknown, all=False, plot_type='ROC')" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "2026749f1caa1790c4f6e7a2adfa1a5f", "grade": true, "grade_id": "knn_classifier_test", "locked": true, "points": 3, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# Test for grading\n", "knn = Knn_Classifier(X_unknown, X_labeled, y_labeled)\n", "res = np.asarray([3, 8, 3, 3, 8, 9, 5, 5, 2, 9, 6, 7, 2, 2, 7, 8, 0, 1, 2, 0, 5, 6, 8, 1, 1, 6, 0, 7, 7, 4, 4, 4, 4, 4, 4])\n", "assert np.all(knn.nn_targets[2]==res), 'visible test: something wrong in nn_targets'\n", "\n", "knn.set_k(5)\n", "assert knn.knn_targets.shape[-1]==5, 'visible test: something wrong in knn_targets shape'\n", "\n", "#similar test as above, hidden\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "1d607a5b18e999c142cccaa09e55762c", "grade": true, "grade_id": "binary_metrics_test", "locked": true, "points": 2, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# Test for grading\n", "knn = Knn_Classifier(X_unknown, X_labeled, y_labeled)\n", "knn.set_k(5)\n", "metrics = compute_binary_metrics(knn, label=5, ground_truth=y_unknown, threshold=.3)\n", "res = {'TP': 141, 'FP': 38, 'TN': 1545, 'FN': 38, 'precision': 0.7877094972067039, 'recall': 0.7877094972067039, 'fall-out': 0.024005053695514846}\n", "for k in metrics.keys():\n", " assert metrics[k]==res[k], f'visible test: metric {k} is wrong'\n", "\n", "#similar test as above, hidden\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "ae41c0f0fad9dee6e57ff29c00b1fdc4", "grade": false, "grade_id": "cell-69e2c302f41a7a78", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "### 3.3 Dimensionality reduction with PCA" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "7be164b88d02b756ee692ac8a7df2ea9", "grade": false, "grade_id": "cell-e295194e814491ac", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "Now we will try to reduce the number of features or pixels to two and re-test the precision/recall curves from exercise 3.2.\n", "\n", "We will try with an unsupervised method, namely via Principle Component Analysis (PCA). The task is to use `scikit-learn`'s implementation of the PCA to reduce the number of pixels to two and then to make a scatter plot of the new features, where the color of each point represents the digits number. Which property should this scatterplot have in order for the new features to be especially suitable for similarity search?\n", "\n", "Implement a pca model using the imported sklearn function and transform the image features retaining only 2 components." ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "deletable": false, "nbgrader": { "cell_type": "code", "checksum": "78fa8affbb72cf36befd28cb9b0bf2a5", "grade": false, "grade_id": "PCA", "locked": false, "schema_version": 3, "solution": true, "task": false } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from sklearn.decomposition import PCA\n", "\n", "pca = PCA(n_components=2)\n", "\n", "new_features_pca = pca.fit_transform(X_unknown)\n", "\n", "plt.figure(figsize=(9, 7))\n", "for i in range(10):\n", " plt.scatter(new_features_pca [y_unknown==i, 0], new_features_pca [y_unknown==i, 1], label=str(i))\n", "plt.xlabel('Dim 1')\n", "plt.ylabel('Dim 2')\n", "plt.title('2D Feature Space')\n", "plt.legend()\n", "plt.show()\n", "plt.close()" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "code", "checksum": "fc3cb02ef9ab49135c0eb122bd44b7e5", "grade": true, "grade_id": "PCA_test", "locked": true, "points": 1, "schema_version": 3, "solution": false, "task": false } }, "outputs": [], "source": [ "##### DO NOT CHANGE #####\n", "\n", "# Test for grading\n", "#checking if correct number of components is retained in new_features_pca\n", "\n", "##### DO NOT CHANGE #####" ] }, { "cell_type": "markdown", "metadata": { "deletable": false, "editable": false, "nbgrader": { "cell_type": "markdown", "checksum": "c82b6a4ef44b3c03a7386a2cee3d3e3c", "grade": false, "grade_id": "cell-41170211ff7fdd39", "locked": true, "schema_version": 3, "solution": false, "task": false } }, "source": [ "We fixed the number of components to be retained by the PCA to 2, so that we could easily visualize them with the scatter plot. What would be more sensible is to select a number of components sufficient enough to represent the variance of the data, in such a way to keep almost all valuable information carried by our features, but hopefully smaller than the total number of features, so to avoid redundancy. \n", "\n", "That's what we are doing next." ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Original number of features: 64\n", "Using only 12 dimensions is enough to explain the 90% of the variance in the data\n" ] } ], "source": [ "pca = PCA()\n", "pca.fit(X_labeled)\n", "explained_variance = np.cumsum(pca.explained_variance_ratio_)\n", "n_dim = next(ind for ind, x in enumerate(explained_variance) if x > .9)\n", "print(f'Original number of features: {X_labeled.shape[-1]}')\n", "print(f'Using only {n_dim} dimensions is enough to explain the 90% of the variance in the data')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's plot now the ROC curves of the knn classifier trained using all features and of the knn classifier with PCA reduced features. The performance improves only slightly in this specific example, but PCA reduction is a useful method to avoid overfitting in settings where the dimension of the dataset is small with respect to the dimension of the feature space." ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "ROC curve of knn classifier using all available features\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkIAAAHHCAYAAABTMjf2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABp5UlEQVR4nO3dd3xT9f7H8VfSXWgptIyCQAFBkD0sIqCgKIigqPeCoKyLoAKuOhAHQ72AiohXUa4oQ0XBhfJTRFm9IqAgUBdLGbJHGS10ps35/XFISqVAW5qepHk/H48+JCfnJJ/0FPvmO22GYRiIiIiI+CG71QWIiIiIWEVBSERERPyWgpCIiIj4LQUhERER8VsKQiIiIuK3FIRERETEbykIiYiIiN9SEBIRERG/pSAkIiIifktBSERERPyWgpCID5s9ezY2m839FRgYSI0aNRg0aBD79u0r8BrDMHjvvfe4+uqriYqKIjw8nKZNm/Lss8+SlpZ2zvdasGABN954IzExMQQHB1O9enV69+7N8uXLL/pz9O7dG5vNxqhRo877OX/66acCn+/RowdxcXFnHc/MzOSVV16hbdu2VKhQgdDQUBo0aMDIkSPZtm1bses9ceIEw4YNo3LlypQrV47OnTuzYcOGQl+/efNmunXrRvny5alUqRL9+/fnyJEj+c7ZtWtXvnt75te8efPynTto0KACz2vYsGGxP6OIvwi0ugARuXjPPvssderUITMzkx9++IHZs2fz/fff89tvvxEaGuo+Lzc3l379+vHRRx/RsWNHxo0bR3h4OCtXrmT8+PF8/PHHLF26lKpVq7qvMQyDf/3rX8yePZuWLVuSkJBAtWrVOHDgAAsWLOC6665j1apVXHXVVcWqPTU1lf/7v/8jLi6ODz/8kEmTJmGz2S76e5KcnEy3bt1Yv349PXr0oF+/fpQvX56tW7cyb9483nrrLbKzs4v8uk6nk5tuuomff/6Zxx57jJiYGN544w06derE+vXrqV+//nmv37t3L1dffTUVKlRgwoQJnDp1ismTJ/Prr7+ydu1agoOD853ft29funfvnu9Yu3btznrdkJAQ3n777XzHKlSoUOTPJ+J3DBHxWbNmzTIAY926dfmOjxo1ygCM+fPn5zs+YcIEAzAeffTRs15r4cKFht1uN7p165bv+EsvvWQAxkMPPWQ4nc6zrnv33XeNH3/8sdifYebMmUZQUJCxfPlyAzASExPPOudcn9PlpptuMmrXrn3WMbvdbnzyySdnnZ+ZmWk88sgjxap3/vz5BmB8/PHH7mOHDx82oqKijL59+17w+vvuu88ICwsz/vrrL/exJUuWGIDx3//+131s586dBmC89NJLF3zNgQMHGuXKlSviJxERwzAMdY2JlEEdO3YEYPv27e5jGRkZvPTSSzRo0ICJEyeedU3Pnj0ZOHAgixcv5ocffnBfM3HiRBo2bMjkyZMLbKnp378/8fHxxa517ty5XH/99XTu3JlGjRoxd+7cYr+Wy48//shXX33FkCFDuP322896PiQkhMmTJ7sfOxwOtmzZwoEDBy742p988glVq1bltttucx+rXLkyvXv35osvviArK+u813/66af06NGDWrVquY916dKFBg0a8NFHHxV4TVpaWqFar3Jzc0lNTb3geSKSR0FIpAzatWsXABUrVnQf+/777zl+/Dj9+vUjMLDgXvEBAwYA8OWXX7qvOXbsGP369SMgIKDE69y/fz8rVqygb9++gNkN9MknnxSry+pMCxcuBMyQVhj79u2jUaNGjB49+oLnbty4kVatWmG35//fZ3x8POnp6ecde7Rv3z4OHz5MmzZtznouPj6ejRs3nnV8/PjxlC9fntDQUK644gq+/fbbAl87PT2dyMhIKlSoQKVKlRgxYgSnTp264OcR8XcKQiJlQEpKCsnJyezdu5dPP/2U8ePHExISQo8ePdznbNq0CYDmzZuf83Vcz23evDnff5s2beqRuj/88ENCQkK45ZZbALjjjjs4fvw4ixYtuqjX9WTdBw4cIDY29qzjrmP79+8/77Vnnvv3648dO+ZuUbLb7dxwww289NJLLFy4kFdeeYXDhw9z44038tVXX5117eOPP86sWbP48MMPufnmm3njjTfo1q0bOTk5xf6sIv5Ag6VFyoAuXbrkexwXF8f777/PJZdc4j528uRJACIiIs75Oq7nXN0rrv+e75qLMXfuXG666Sb369evX5/WrVszd+5cevXqVezXLWrdcXFxGIZRqHMzMjIICQk567hrUHpGRsZ5rwUueH1ISAi1atXim2++yXdO//79ufzyy3nkkUe46aab3Mf/3tV5xx130KBBA5566ik++eQT7rjjjkJ9NhF/pBYhkTJg2rRpLFmyhE8++YTu3buTnJx81i9bVyhwBaKC/D0sRUZGXvCa4tq8eTMbN26kffv2/Pnnn+6vTp068eWXXxZ5rMuZ45c8WXdYWFiB44AyMzPdz5/vWqDY11eqVInBgwezdetW9u7de946H374Yex2O0uXLj3veSL+TkFIpAyIj4+nS5cu3H777SxcuJAmTZrQr1+/fGNEGjVqBMAvv/xyztdxPXf55ZcDuNeh+fXXX0u85vfffx8wf2HXr1/f/fXyyy+TmZnJp59+6j73Qq0t6enp+ZYJ8GTdsbGxBQ6qdh2rXr36ea8989y/X1+pUqUCW4vOVLNmTQCOHTt23vPCwsKIjo6+4Hki/k5BSKSMCQgIYOLEiezfv5/XX3/dfbxDhw5ERUXxwQcfkJubW+C17777LoB7bFGHDh2oWLEiH3744TmvKQ7DMPjggw/o3LkzH3/88VlfzZo1yzd7rHbt2gBs3bq1wNfbtm2b+xwwZ8BBXtgqSS1atGDDhg04nc58x3/88UfCw8Np0KDBOa+tUaMGlStXLnBhyLVr19KiRYsLvv+OHTsAc6ba+Zw8eZLk5OQLnifi96yevy8ixXe+9XXi4+ONqlWrGhkZGe5jzz//vAEYo0aNOuv8L7/80rDb7UbXrl3zHZ80aZIBGI888kiB6wi99957RV5HaOXKlQZgvPvuuwU+/+9//9uw2+3Gvn37DMMwjKysLKNKlSpG27ZtjczMzHznLliwwACMqVOn5jverVs3w263GwsWLDjr9bOysvKtI5SdnW1s3rzZ2L9//wVrnzdv3lnrCB05csSIiooy+vTpk+/cP//80/jzzz/zHbv33nuNsLAwY/fu3e5jS5cuNQDjzTffdB87fPjwWe+9d+9eo2LFikazZs3cxzIyMozU1NSzzn3ssccMwPjss88u+JlE/JmCkIgPO18Q+vjjj8/65ZqTk2PcfvvtBmBcffXVxquvvmq89dZbxoABAwy73W40btzYOHjwYL7Xyc3NNfr3728ARqtWrYwJEyYYM2fONCZMmGDEx8cbgLF69Wr3+bVr1z5rccO/u/fee42AgADj6NGjBT7/66+/GoDx8ssvu4+9++67hs1mM5o2bWo8//zzxptvvmkMGzbMCAwMNJo1a2akp6fne43Dhw8bLVq0MGw2m3HzzTcbr776qvH2228bo0aNMmrXrm0EBwe7z3UtXjhw4MDz1u36Hl555ZVG+fLljfHjxxvTpk0zGjdubERERBhbtmzJd25B34vdu3cb0dHRRr169Yz//Oc/xoQJE4yKFSsaTZs2zRfyBg0aZHTs2NEYN26c8dZbbxlPPvmkER0dbQQHBxsrVqzIV3tUVJRx3333Ga+++qrx6quvGt27dzcAo1u3bkZubu4FP5OIP1MQEvFh5wtCubm5Rr169Yx69eoZOTk5+Y7PmjXLaN++vREZGWmEhoYajRs3NsaPH2+cOnXqnO/1ySefGDfccINRqVIlIzAw0IiNjTX69Olz1krQMTExxpVXXnnO18nOzjaio6ONjh07nvez1alTx2jZsmW+Y19//bXRuXNnIzIy0ggKCjLq1KljJCQkGMePHy/wNdLT043JkycbV1xxhVG+fHkjODjYqF+/vnH//ffna6kpShAyDMM4duyYMWTIECM6OtoIDw83rrnmmgLvwblC4W+//WbccMMNRnh4uBEVFWXceeedZwXQDz74wLj66quNypUrG4GBgUZMTIxx6623GuvXr8933vHjx4277rrLuPTSS43w8HAjJCTEaNy4sTFhwgQjOzu7UJ9HxJ/ZDKOQc0ZFRC5g06ZNNG7cmC+//DLf9G4REW+lwdIiUmJWrFhBu3btFIJExGeoRUhERET8llqERERExG9ZGoS+++47evbsSfXq1bHZbHz++ecXvCYxMZFWrVoREhLCpZdeyuzZsz1ep4iIiJRNlgahtLQ0mjdvzrRp0wp1/s6dO7npppvo3LkzSUlJPPTQQ9x9991n7ccjIiIiUhheM0bIZrOxYMGC8260OGrUKL766it+++0397E77riDEydOsHjx4lKoUkRERMoSn9p9fs2aNWftst21a1ceeuihc16TlZWVb4NDp9PJsWPHiI6OzrdJo4iIiHgvwzA4efIk1atXx24vuQ4tnwpCBw8epGrVqvmOVa1aldTUVDIyMgrctXnixImMHz++tEoUERERD9qzZw+XXHJJib2eTwWh4hg9ejQJCQnuxykpKdSqVYtt27ZRqVIlCysTh8PBihUr6Ny5M0FBQVaX4/d0P7yH7oX30L2wlm3hQowuXSA8nGPHjtGgQQMiIiJK9D18KghVq1aNQ4cO5Tt26NAhIiMjC2wNAggJCSEkJOSs45UqVSI6OtojdUrhOBwOwsPDiY6O1v9gvIDuh/fQvfAeuhcWSUuDESNgzhy4+26YMcP9VEkPa/GpINSuXTsWLVqU79iSJUto166dRRWJiIhIifrtN+jdGzZvBrsdatUCD87rsnT6/KlTp0hKSiIpKQkwp8cnJSWxe/duwOzWGjBggPv8e++9lx07dvD444+zZcsW3njjDT766CMefvhhK8oXERGRkmIY8M47EB9vhqDYWFi2DJ55Bjw4ucnSIPTTTz/RsmVLWrZsCUBCQgItW7ZkzJgxABw4cMAdigDq1KnDV199xZIlS2jevDkvv/wyb7/9Nl27drWkfhERESkBp05B//5mN1hGBtxwAyQlQadOHn9rS7vGOnXqxPmWMSpo1ehOnTqxceNGD1Zlys3NxeFwePx9/JnD4SAwMJDMzExyc3OtLsejgoKCCAgIsLoMERHvdOIELF4MAQHw/PPw+ONmt1gp8KkxQqXBMAwOHjzIiRMnrC6lzDMMg2rVqrFnzx6/WNMpKiqKatWq+cVnFREpkksugQ8/hLAw6NChVN9aQehvXCGoSpUqhIeH65eWBzmdTk6dOkX58uVLdHEsb2MYBunp6Rw+fBiA2NhYiysSEbFYaioMGwZ33AGuHSWuv96SUhSEzpCbm+sOQZpa73lOp5Ps7GxCQ0PLdBAC3Ms7HD58mCpVqqibTET81/r10KcPbN8OK1aY44HCwy0rp2z/9iki15igcAtviJRdrp8rjT0TEb9kGPDaa3DVVWYIql0bvvjC0hAEahEqkLrDxBP0cyUifuvECRgyBD77zHzcqxfMnAkVK1pZFaAgJCIiIp504gS0bAm7dkFQEEyeDPff79G1gYpCXWNyTuPGjaNFixbnPWfXrl3YbDb3opgXMmjQIHq5BsaJiEjZFxUFN94IdevC6tXwwANeE4JALUJSBIMGDeLEiRN8/vnn7mM1a9bkwIEDxMTEWFeYiIh4l6NHIScHqlY1H0+ZAllZUKGCtXUVQC1CclECAgKoVq0agYHK1CIigtnq07Il9O0LrsVyQ0O9MgSBglCZ0alTJ+6//34eeughKlasSNWqVZkxYwZpaWkMHjyYiIgILr30Ur7++mvAXLU7Kioq32t8/vnn5xzQO27cOObMmcMXX3yBzWbDZrORmJhYYNfY77//To8ePYiMjCQiIoKOHTuyffv2Al936dKlXH311URFRREdHU2PHj3ynZudnc3IkSOJjY0lNDSU2rVrM3HiRMBcn2fcuHHUqlWLkJAQqlevzgMPPHAR30URESk2pxNeeAGuvhr27DG/DhywuqoL0j/jL8AwDDIc1mz/EBYUUKSZRnPmzOHxxx9n7dq1zJ8/n/vuu48FCxZw66238uSTT/LKK6/Qv3//fPu3Fdajjz7K5s2bSU1NZdasWQBUqlSJ/fv35ztv3759XH311XTq1Inly5cTGRnJqlWryMnJKfB109PTeeihh2jRogWnTp1izJgx3HrrrSQlJWG32/nPf/7DwoUL+eijj6hVqxZ79uxhz549AHz66ae88sorzJs3j8aNG3Pw4EF+/vnnIn82ERG5SEeOwMCBcPof2/TtC//9L0REWFtXISgIXUCGI5fLx3xjyXtverYr4cGFv0XNmzfn6aefBmD06NFMmjSJmJgYhg4dCsCYMWN48803+eWXX4pcS/ny5QkLCyMrK4tq1aqd87xp06ZRoUIF5s2bR1BQEAANGjQ45/k333wzkZGR7gUVZ86cSeXKldm0aRNNmjRh9+7d1K9fnw4dOmCz2ahdu7b72t27d1OtWjW6dOlCUFAQtWrVIj4+vsifTURELsLKleYK0fv3m11g//mPuXmqFw2IPh91jZUhzZo1c/85ICCA6OhomjZt6j5W9fSgNddWD56QlJREx44d3SHoQrZv306/fv2oW7cukZGRxMXFAbhbrQYNGkRSUhKXXXYZDzzwAN9++6372n/+859kZGRQt25dhg4dyoIFC87Z8iQiIh6QmwvDh5shqGFDWLsWhg71mRAEahG6oLCgADY929Wy9y6Kv4cPm82W75irm83pdGK32zEMI9/5JbHisWsricLq27cvcXFxzJgxg+rVq+N0OmnSpAnZ2dkAtGrVip07d/L111+zdOlSevfuTZcuXfjkk0+oWbMmW7duZenSpSxZsoThw4fz0ksv8b///a/QQUxERC5CQIC5Weqrr8Irr0D58lZXVGQKQhdgs9mK1D3lKypXrszJkydJS0ujXLlyABdcCyg4OJjc3POPl2rWrBlz5szB4XBcMIwcPXqUP/74gxkzZnDNNdcA8P333591XmRkJH369KFPnz784x//oFu3bhw7doxKlSoRFhZGz5496dmzJyNGjKBhw4b8+uuvtGrV6rzvLSIixbR8OfzxB9xzj/m4SROYMcPami5C2fsNL4XStm1bwsPDefLJJ3nggQf48ccfmT179nmviYuL45tvvmHr1q1ER0dToYCpkCNHjuS1117jjjvuYPTo0VSoUIEffviB+Ph4LrvssnznVqxYkUqVKjFjxgxq1KjB7t27eeKJJ/KdM2XKFGJjY2nZsiV2u52PP/6YatWqERUVxezZs8nNzXV/lvfff5+wsLB844hERKSE5ObCs8/Cc8+ZLUGtW0ObNlZXddE0RshPVapUiffff59FixbRtGlTPvzwQ8aNG3fea4YOHcpll11GmzZtqFy5MqtWrTrrnOjoaJYvX86pU6e45ppraN26NTNmzCiwdchut/POO++wYcMGmjRpwsMPP8xLL72U75yIiAhefPFF2rRpwxVXXMGuXbtYtGgRdrudqKgoZsyYQfv27WnWrBlLly7l//7v/4iOjr6o742IiPzN/v3QpYsZhAwDBg2Cyy+3uqoSYTP+PlCkjEtNTaVChQokJyef9QszMzOTnTt3UqdOHUJDQy2q0H84nU5SU1PzzRory7z958vhcLBo0SK6d++uMVYW073wHroXwDffQP/+5hT58uXNafH9+pV6GUePHiUmJoaUlBQiIyNL7HXL/m8fERERKZ5x46BbNzMENW8O69dbEoI8SUFIRERECubageDee+GHH+A868L5Kg2WFhERkTxpaXB6NjEPPmjuG3Z6Zm9ZpBYhERERAYcDHnsMWrWCkyfNYzZbmQ5BoCAkIiIif/1lbpY6eTJs2waff251RaVGQUhERMSfffEFtGhhjgGqUAE+/dScJeYnFIRERET8UXY2PPQQ9OoFJ05AfDxs3Ai33WZxYaVLQUhERMQfjRpl7hEG8Mgj5i7ydepYW5MFFIRERET80RNPQOPGsHChOTYoONjqiiyhICRFNmjQIHr16mV1GSIiUhSZmeZO8S5Vq8Ivv0DPntbV5AW0jpCIiEhZ98cf0Ls3JCWZj/v2Nf/rB9sbXYi+A2VUdna21SWIiIg3+PBDc22gpCSIiYFKlayuyKsoCJURnTp1YuTIkTz00EPExMTQtWtXpkyZQtOmTSlXrhw1a9Zk+PDhnDp1yn3N7NmziYqK4ptvvqFRo0aUL1+ebt26ceDAAfc5ubm5JCQkEBUVRXR0NI8//jh/36c3KyuLBx54gCpVqhAaGkqHDh1Yt26d+/nExERsNhvffPMNLVu2JCwsjGuvvZbDhw+zZMkSGjduTGRkJP369SM9Pb1Qn/fkyZPceeedlCtXjtjYWF555RU6derEQw895D7HZrPx+d/WwoiKimL27Nnux3v27KF3795ERUVRqVIlbrnlFnbt2pWv9vj4eMqVK0dUVBTt27fnr7/+AuDnn3+mc+fOREREEBkZSevWrfnpp58KVb+IiMdlZMCwYebeYKdOmesEJSVB165WV+ZVFIQKKy3t3F+ZmYU/NyOjcOcWw5w5cwgODmbVqlVMnz4du93Of/7zH37//XfmzJnD8uXLefzxx/Ndk56ezuTJk3nvvff47rvv2L17N48++qj7+ZdffpnZs2czc+ZMvv/+e44dO8aCBQvyvcbjjz/Op59+ypw5c9iwYQOXXnopXbt25dixY/nOGzduHK+//jqrV69mz5493HHHHUyfPp3333+fr776im+//ZbXXnutUJ81ISGBVatWsXDhQpYsWcLKlSvZsGFDkb5fDoeDrl27EhERwcqVK1m1apU7DGZnZ5OTk0OvXr245ppr+OWXX1izZg3Dhg3DZrMBcOedd3LJJZewbt061q9fzxNPPOG/u1OLiHfZsgXatoUZM8zVoZ9+GpYtgxo1rK7M+xh+JiUlxQCM5OTks57LyMgwNm3aZGRkZJx9IZz7q3v3/OeGh5/73GuuyX9uTEzB5xXRNddcY7Rs2fK853z88cdGdHS0+/GsWbMMwPjzzz/dx6ZNm2ZUrVrV/Tg2NtZ48cUX3Y8dDodxySWXGLfccothGIZx6tQpIygoyJg7d677nOzsbKN69eru61asWGEAxtKlS93nTJw40QCMjRs3Grm5uYZhGMY999xjdO3a9YKfNTU11QgKCjI+/vhj97ETJ04Y4eHhxoMPPug+BhgLFizId22FChWMWbNmGYZhGO+9955x2WWXGU6n0/18VlaWERYWZnzzzTfG0aNHDcBITEwssI6IiAhj9uzZF6zX5bw/X14gOzvb+Pzzz43s7GyrS/F7uhfew2fvxZdfmr9LqlQxjCVLrK6mRCQnJxuAkZKSUqKvqxahMqR169b5Hi9dupTrrruOGjVqEBERQf/+/Tl69Gi+7qfw8HDq1avnfhwbG8vhw4cBSElJ4cCBA7Rt29b9fGBgIG3atHE/3r59Ow6Hg/bt27uPBQUFER8fz+bNm/PV06xZM/efq1atSnh4OHFxcfmOud77fHbs2IHD4SA+Pt59rEKFClx22WUXvPZMP//8M3/++ScRERGUL1+e8uXLU6lSJTIzM9m+fTuVKlVi0KBBdO3alZ49e/Lqq6/m6zZMSEjg7rvvpkuXLkyaNInt27cX6f1FRDzmppvM1qCkJOjSxepqvJqCUGGdOnXur08/zX/u4cPnPvfrr/Ofu2tXwecVQznXbsHArl276NGjB82aNePTTz9l/fr1TJs2Dcg/kPrvXTk2m+2sMUAl5cz3stlsBb630+kssfcr6LM4HA73n0+dOkXr1q1JSkrK97Vt2zb69esHwKxZs1izZg1XXXUV8+fPp0GDBvzwww+A2dX3+++/c9NNN7F8+XIuv/zys7oNRURKxe+/Q8eO5p5hLnffDbGx1tXkIxSECqtcuXN/hYYW/tywsMKde5HWr1+P0+nk5Zdf5sorr6RBgwbs37+/SK9RoUIFYmNj+fHHH93HcnJyWL9+vftxvXr13OOSXBwOB+vWrePyyy+/6M9RkLp16xIUFJRvQHZKSgrbtm3Ld17lypXzteD88ccf+VrDWrVqxR9//EGVKlW49NJL831VqFDBfV7Lli0ZPXo0q1evpkmTJnzwwQfu5xo0aMDDDz/Mt99+y2233casWbM88ZFFRApmGDBzJlxxBXz/vbllhhSJglAZdemll+JwOHjttdfYsWMH7733HtOnTy/y6zz44INMmjSJzz//nC1btjB8+HBOnDjhfr5cuXLcd999PPbYYyxevJhNmzYxdOhQ0tPTGTJkSAl+ojwREREMHDiQxx57jBUrVvD7778zZMgQ7Ha7eyAzwLXXXsvrr7/Oxo0b+emnn7j33nvztULdeeedxMTEcMstt7By5Up27txJYmIiDzzwAHv37mXnzp2MHj2aNWvW8Ndff/Htt9/yxx9/0KhRIzIyMhg5ciSJiYn89ddfrFq1inXr1tGoUSOPfGYRkbOcOmVujjpkiDkR54Yb4L//tboqn6MFFcuo5s2bM2XKFF544QVGjx7N1VdfzcSJExkwYECRXueRRx7hwIEDDBw4ELvdzr/+9S9uvfVWUlJS3OdMmjQJp9NJ//79OXnyJG3atOGbb76hYsWKJf2x3KZMmcK9995Ljx49iIyM5PHHH2fPnj2EntE69/LLLzN48GA6duxI9erVefXVV/O1ZoWHh/Pdd98xatQobrvtNk6ePEmNGjW47rrriIyMJCMjgy1btjBnzhyOHj1KbGwsI0aM4J577iEnJ4ejR48yYMAADh06RExMDLfddhvjx4/32GcWEXH7+WdzgcRt2yAgAJ57ztw7TAskFpnN8NSAEC+VmppKhQoVSE5OJjo6Ot9zmZmZ7Ny5kzp16uT7hSqe4XQ6SU1NJTIyEvtF/uVNS0ujRo0avPzyyx5ribpY3v7z5XA4WLRoEd27d9cyABbTvfAeXnkvVq6E66+HrCxzOvy8edChg9VVedzRo0eJiYkhJSWFyMjIEntdtQiJT9q4cSNbtmwhPj6elJQUnn32WQBuueUWiysTEfGwK66Ahg3NEDRnjrlatBSbgpB4nd27d593oPWmTZsAmDx5Mlu3biU4OJjWrVuzcuVKYvQ/BBEpizZvhgYNzG6w0FBYutTcKkNdYRdNQUi8TvXq1UlybQx4judr1aqVb7yPiEiZZBgwbRo88gg89RSMGWMe1z/6SoyCkHidwMBALr30UqvLEBGx1okT5oywzz4zH//8MzidagUqYfpuFsDPxo9LKdHPlYgU2tq10LKlGYKCgmDqVPjkE4UgD9B39AyuGQGF3QFdpChcP1deM/NERLyPYcArr5izwHbtgjp1YNUqePBBc/NUKXHqGjtDQEAAUVFR7v2uwsPD8y3QJyXL6XSSnZ1NZmbmRU+f92aGYZCens7hw4eJiooiICDA6pJExFvt3AlPPgkOB9x+O7z9NkRFWV1VmaYg9DfVqlUDKNTmn3JxDMMgIyODsLAwvwicUVFR7p8vEZEC1a1rDo7OyIDhw9UKVAoUhP7GZrMRGxtLlSpV8m3QKSXP4XDw3XffcfXVV5f57qKgoCC1BInI2ZxOePllc8PUK680j/3rX9bW5GcUhM4hICBAv7g8LCAggJycHEJDQ8t8EBIROcuRIzBwIHz9NdSuDb/9BuXLW12V31EQEhERKW3ffQd9+8L+/eYCiU89BeXKWV2VXyq7I1RFRES8jdMJ//43dO5shqDLLoMff4ShQzUeyCJqERIRESkNp07BbbfBkiXm4/794Y031B1mMQUhERGR0lCuHISFmV9vvAGDBlldkaAgJCIi4jm5uZCdbYYfmw1mzYKDB+E8G0tL6dIYIREREU84cAC6dDHH/7i22KlUSSHIy6hFSEREpKR9+y3cdZc5Rb5cOdixA+rVs7oqKYBahEREREpKTo45Fb5bNzMENWsGP/2kEOTF1CIkIiJSEvbuhX79YOVK8/E995gbqIaFWVuXnJeCkIiIyMVyOuHGG83VoSMiYMYM6NPH6qqkENQ1JiIicrHsdpg6Fdq0gQ0bFIJ8iIKQiIhIcezebQ6KdrnuOnOV6Esvta4mKTIFIRERkaJauBBatIB//AP+/DPvuF2/Vn2N7piIiEhhZWfDww/DLbfA8ePQsCEEaritL7M8CE2bNo24uDhCQ0Np27Yta9euPe/5U6dO5bLLLiMsLIyaNWvy8MMPk5mZWUrVioiI39q5Ezp0MMcCgRmIvv8e4uKsrEoukqVBaP78+SQkJDB27Fg2bNhA8+bN6dq1K4cPHy7w/A8++IAnnniCsWPHsnnzZt555x3mz5/Pk08+WcqVi4iIP7F99hm0bAnr1kHFivDFFzBlCgQHW12aXCRLg9CUKVMYOnQogwcP5vLLL2f69OmEh4czc+bMAs9fvXo17du3p1+/fsTFxXHDDTfQt2/fC7YiiYiIXAzbDz9ASgq0awdJSXDzzVaXJCXEso7N7Oxs1q9fz+jRo93H7HY7Xbp0Yc2aNQVec9VVV/H++++zdu1a4uPj2bFjB4sWLaJ///7nfJ+srCyysrLcj1NTUwFwOBw4HI4S+jRSHK7vv+6Dd9D98B66F17CMHDk5ACQNXYsIbVq4bznHggKAt2bUuepvw+WBaHk5GRyc3OpWrVqvuNVq1Zly5YtBV7Tr18/kpOT6dChA4ZhkJOTw7333nverrGJEycyfvz4s46vWLGC8PDwi/sQUiKWLFlidQlyBt0P76F7YZ0aK1dSc8UKfnzySQgMZMn//gd164LuiWXS09M98ro+NdQ9MTGRCRMm8MYbb9C2bVv+/PNPHnzwQZ577jmeeeaZAq8ZPXo0CQkJ7sepqanUrFmTzp07Ex0dXVqlSwEcDgdLlizh+uuvJygoyOpy/J7uh/fQvbBQRgb2Rx4h4O23Aei2Zw9f16mje+EFjh496pHXtSwIxcTEEBAQwKFDh/IdP3ToENWqVSvwmmeeeYb+/ftz9913A9C0aVPS0tIYNmwYTz31FPYC1m8ICQkhJCTkrONBQUH6ofYSuhfeRffDe+helLKtW6F3b/jlF7DZ4Mknsd1zD3z7re6FF/DU99+ywdLBwcG0bt2aZcuWuY85nU6WLVtGu3btCrwmPT39rLATEBAAgGEYnitWRETKtvffh9atzRBUpQp88w08/7zWCPIDlt7hhIQEBg4cSJs2bYiPj2fq1KmkpaUxePBgAAYMGECNGjWYOHEiAD179mTKlCm0bNnS3TX2zDPP0LNnT3cgEhERKZJ//xueftr8c+fOMHcuxMZaW5OUGkuDUJ8+fThy5Ahjxozh4MGDtGjRgsWLF7sHUO/evTtfC9DTTz+NzWbj6aefZt++fVSuXJmePXvy73//26qPICIivu4f/4AXX4SEBDMQ6R/WfsXyNr+RI0cycuTIAp9LTEzM9zgwMJCxY8cyduzYUqhMRETKJMMwu8CaNzcfX3YZ7NgBmkDjlyzfYqMs2nHkFNk5TqvLEBGRvzt1CgYMgFat4H//yzuuEOS3FIRK2Je/7Ofal//Hq8u2WV2KiIic6ZdfoE0bc2A0wG+/WVuPeAUFoRL22YZ9APy445jFlYiICGB2hb31FsTHm1Pka9SAxEQYMcLqysQLWD5GqCxJz85h1Z/JAOw6mmZxNSIiQmoq3HMPzJtnPr7xRnj3XYiJsbYu8RpqESpB3/+RTNbpsUHJp7JJzdReNCIilvriCzMEBQSYM8O+/FIhSPJRi1AJWrb5cL7Hu5LTaHZJlDXFiIgI3HUXbNwI//ynuXO8yN+oRaiEOJ0Gy7aYQSgsyFyDYmeyusdERErViRMwciQcP24+ttlgyhSFIDknBaES8su+FJJPZVE+JJAbm5h7pSkIiYiUonXrzGnx06bBffdZXY34CAWhErJss7l57NUNYqhfNQIwu8ZERMTDDAOmToX27WHnTqhTBx55xOqqxEdojFAJWXp6fNB1DatSLsT8tqpFSETEw44dg8GDYeFC8/Htt8Pbb0NUlKVlie9QECoBe4+ns/lAKnYbdG5YhSMnswAzCBmGgc1ms7hCEZEy6NdfoUcP2L0bgoPNsUDDh5vjgkQKSUGoBCw/PUi6de2KVCoXTHiwOVg6NTOHY2nZRJcPsbI8EZGyqXp1s1usXj346CNzfJBIEWmMUAlwd4s1qgpAaFAANaLCAC2sKCJSok6eNMMPmPuDff01bNigECTFpiB0kU5l5fDD9qMAdGlUxX08LiYcgB1HFIRERErEypXQqBHMnp13rHFjiIy0rCTxfQpCF+n7P46QneukdnQ49SqXdx+vE1MOUIuQiMhFczphwgTo3Bn27YPXXoPcXKurkjJCQeginTlb7MxB0XHRZhDSzDERkYtw+DB06wZPPWWGn7vugu++M7fMECkBGix9EXKdhnug9JndYgB1K7uCUHqp1yUiUiasWAH9+sHBgxAWBq+/bk6V16wwKUEKQhchac9xjqVlExEayBV1KuV7ztUitEtT6EVEiu6vv+CGGyAnBy6/3JwV1rix1VVJGaQgdBFc3WKdLqtCUED+XsaalcIJsNvIcORyKDWLahVCrShRRMQ31a4No0fD3r3mmKBy5ayuSMooBaGL4NpW4+/dYgBBAXZqVgxj19F0dianKQiJiFzI0qUQFweXXmo+Hj9e3WDicRosXUx7jqWz7dApAuw2OjU4OwgBxMVowLSIyAXl5MDTT5tdYX36QJa5Or9CkJQGBaFiWnq6NahN7YpUCA8q8BxNoRcRuYB9++Daa+Hf/zYXSrziirwFE0VKgbrGimmpu1us6jnPcQUhLaooIlKAr7+GAQMgORkiIuCtt+COO6yuSvyMWoSKITXTwY87jgHQ5fILByG1CImInMHhgFGjoHt3MwS1bAnr1ysEiSUUhIrhu21HyHEa1K1czh12CuKaQr/7aDq5TjX1iogAZtfXihXmn0eMgNWroX59a2sSv6WusWJYttm1iOK5W4MAqkeFERxoJzvHyf4TGdSsFF4a5YmIeCfDMAdABwfD/PnmZqm33251VeLn1CJURDm5TlZsdW2rUfBsMZcAu43ap8OPZo6JiN/KzoaEBHObDJc6dRSCxCsoCBXRht0nOJHuoEJYEK1rV7zg+XU0hV5E/NnOndCxI7zyCkyaBFu2WF2RSD4KQkXkmi3W+bLKBAZc+NunICQifuuzz8yB0GvXQlQULFgADRtaXZVIPgpCReSeNn+e2WJn0qKKIuJ3srLg/vvNrq+UFLjySkhKgltusboykbNosHQR7ExOY8eRNALtNq5uULlQ12gKvYj4FcMwV4j+7jvz8eOPw/PPQ1DBC8+KWE1BqAhce4u1rVuJyNDC/aV2BaE9x9LJznESHKhGOBEpw2w2uPtu+P13ePddc60gES+m38pF4OoWu65h4brFAKpEhBAeHIDTgD3H0z1VmoiIdTIyYPPmvMf9+8O2bQpB4hMUhAopJd3Bul3HgQuvH3Qmm83mXlhxp7baEJGyZutWcwxQly5w5Eje8UqVrKtJpAgUhAopcdthcp0G9auUp1Z00RZGrFNZ44REpAx6/31o3Rp++cXcNmPnTqsrEikyBaFCWnp6NenritAa5FInWjPHRKQMSU+HIUPMLrC0NOjUyZwVFh9vdWUiRaYgVAiOXCeJp1eTvv7y868mXRCtJSQiZcamTWbgmTnTHBg9diwsXQrVq1tdmUixaNZYIazbdYyTmTlUKhdMi5oXXk3671xrCe1SEBIRX/fCC+aMsGrVYO5cuPZaqysSuSgKQoXg2mS182VVCLDbiny9q0Vof0omGdm5hAUHlGh9IiKl5j//gcBAmDABqhZ9qICIt1HX2AUYhpG3mnSjoneLAVQMD6JCmLnu0F/H1CokIj7k11/hscfMhRIBKlSAd95RCJIyQ0HoArYfOcVfR9MJDrDTsZCrSf+dzWbL22pDU+hFxBcYBsyYYY4HmjzZDD8iZZCC0AW4Zou1rVuJ8iHF70ms6wpCmkIvIt4uNRX69YNhwyAzE268UfuESZmlIHQBrm01ri/kJqvnokUVRcQnbNxorg00bx4EBJiDo7/8EioXr0VcxNtpsPR5HE/LZv1f5mrS1zYs3vggFy2qKCJe7733zH3CsrOhZk0zDF11ldVViXiUWoTOY8XWwzgNaFgtgksqFm016b/LW1RR+42JiJeqUwdyc6FnT3OBRIUg8QNqETqPvNliFz87Ii7GDFLJp7I4mekgopC714uIeFRKijkTDKBDB1izBtq0MRdLFPEDahE6h+wcJ99tSwbgumJOmz9TRGgQMeVDANilViERsZphwKuvQlycuVq0yxVXKASJX1EQOocfdx7lVFYOMeVDaH5JVIm8Zp3TrUI7kk+VyOuJiBTLsWNw663w0ENw4gTMnm1xQSLWURA6B9dq0tc1rIK9GKtJF6SOe6sNtQiJiEV++AFatoQvvoDgYHjtNXNmmIifUhAqwJmrSZdEt5iLe1FFtQiJSGlzOs2FETt2hN27oV49WL0aRo5UV5j4NQWhAmw7dIq9xzMIDrTToX5Mib1u3qKKahESkVL2/vvmVhk5OdC7N6xfb64XJOLnNGusACv/OALAVfWiCQ8uuW9R3jYbpzAMA5v+FSYipaVfP3O3+FtvhXvuUSuQyGlqESpA8qlsAOpVLl+ir+taXTo1M4fj6Y4SfW0RkXycTnj7bcjKMh8HBsLixXDvvQpBImdQECpAaqYZUiJLeK2f0KAAqlcIBWBnslaYFhEPOXzY3B9s6FAYNSrvuAKQyFkUhAqQmnE6CIWVfM+ha6sNBSER8YjERGjRAr79FsLCoFkzqysS8WoKQgVIzcwBSr5FCPK6x3YpCIlIScrNhWefheuugwMHoFEjWLcO/vUvqysT8WoaLF2AvBahkg9CdWLUIiQiJezgQbjzTli+3Hw8eLC5PlC5ctbWJeIDFIQK4A5CoR7oGlMQEpGSlp4OP/0E4eEwfTr07291RSI+Q0GoAK7B0hXCPdA15lpd+miaptCLSPEZRt7g57p14aOPoHZtaNjQ2rpEfIzGCP2NYRikZnhujFDNiuEE2G2kZ+dy+GRWib++iPiBffvg2mvNAdEuXbsqBIkUg4LQ32TlOMnOdQKeGSMUHGjnkophAOw4ou4xESmixYvNWWGJiTB8uLlStIgUm4LQ37jGB9ltUC44wCPvUeeM7jERkUJxOOCJJ8z1gZKTzTC0aJG5UKKIFJuC0N+4F1MMC/LY+B3XFHoNmBaRQtmzBzp1ytslfvhwWLMGGjSwtCyRssDyIDRt2jTi4uIIDQ2lbdu2rF279rznnzhxghEjRhAbG0tISAgNGjRg0aJFJVZPigfHB7nU1aKKIlJY+/aZrT+rV0NkJHz8MUybBqGhVlcmUiZY2qY6f/58EhISmD59Om3btmXq1Kl07dqVrVu3UqVKlbPOz87O5vrrr6dKlSp88skn1KhRg7/++ouoqKgSq8mTq0q7aFFFESm0GjWgZ0/4/XeYP9+cISYiJcbSIDRlyhSGDh3K4MGDAZg+fTpfffUVM2fO5Iknnjjr/JkzZ3Ls2DFWr15NUJDZYhMXF1eiNXlqn7EzucYI/XU0nVynQYBdU+hF5Ay7dhGcmpr3+I03ICAAQkKsq0mkjLIsCGVnZ7N+/XpGjx7tPma32+nSpQtr1qwp8JqFCxfSrl07RowYwRdffEHlypXp168fo0aNIiCg4IHNWVlZZGXlTVNPPf0/F4fDgcNx9g7wx09lAhARElDg8yWhcrlAggJsZOc62Z180j2LzN+4vr+e+j5L0eh+eAfb558TOHQoLS+9FMett5oHT//DD92bUqe/F97DU/fAsiCUnJxMbm4uVatWzXe8atWqbNmypcBrduzYwfLly7nzzjtZtGgRf/75J8OHD8fhcDB27NgCr5k4cSLjx48/6/iKFSsIDw8/6/i6vTYggJTkgyU69ujvKgUHcCjDxkdfJ9IwyvDY+/iCJUuWWF2CnEH3wxp2h4PGs2dT96uvAAg+eZLEL77AUb68xZUJ6O+FN0hPT/fI6/rUvEun00mVKlV46623CAgIoHXr1uzbt4+XXnrpnEFo9OjRJCQkuB+npqZSs2ZNOnfuTHR09Fnn//rNNtizi8b169C922Ue+yz/d3wjh7YcoUq9xnRvW8tj7+PNHA4HS5Ys4frrr3d3dYp1dD8stH07AXfeiX3DBgAcDz7I9x070uXGG3UvLKa/F97j6NGjHnldy4JQTEwMAQEBHDp0KN/xQ4cOUa1atQKviY2NJSgoKF83WKNGjTh48CDZ2dkEBwefdU1ISAghBfSrBwUFFfhDnZadC0BUeIhHf+jrVomALUf461im3//lOte9EGvofpSyjz6Cu++GkychOhrmzIEbbsBYtEj3wovoXljPU99/y6bPBwcH07p1a5YtW+Y+5nQ6WbZsGe3atSvwmvbt2/Pnn3/idDrdx7Zt20ZsbGyBIag43NtreGBV6TNpUUURITMTRo82Q1D79pCUBDfdZHVVIn7F0nWEEhISmDFjBnPmzGHz5s3cd999pKWluWeRDRgwIN9g6vvuu49jx47x4IMPsm3bNr766ismTJjAiBEjSqymlFKYPg9aVFFEMNcCmj8fnnzS3DLjkkusrkjE71g6RqhPnz4cOXKEMWPGcPDgQVq0aMHixYvdA6h3796N3Z6X1WrWrMk333zDww8/TLNmzahRowYPPvggo0aNKrGaSmP6POQtqrj3eAaOXCdBAZavbSkipeGDDyA93ewOA2jTxvwSEUtYPlh65MiRjBw5ssDnEhMTzzrWrl07fvjhB4/V41pQsYKHu8aqRIQQHhxAenYue46lU7eyZoaIlGnp6fDgg/D22xAcbHaFNWpkdVUifk/NEH+Tmlk6Y4RsNhu11T0m4h82b4a2bc0QZLOZ44K0T5iIV1AQOoNhGHlbbHi4awygboyCkEiZN2eO2fX1229QtSosXQrjxpkrRYuI5SzvGvMmGY5ccpzm4oaeHiwNEBdjLuioICRSBhkGDB0K77xjPu7SBd5/3wxDIuI11CJ0BtfU+UC7jbAgz/9rrU6MOS5IU+hFyiCbzdwg1W6H556DxYsVgkS8kFqEzpA3dT4Im83zG6HWcbUIHVEQEikTDANSUiAqynz8xBPQrRu0amVpWSJybmoROkPe1PnSyYeuFqH9KZlkOnJL5T1FxENOnoQ774SOHc0ZYmC2BikEiXg1BaEzpJ7RIlQaKoYHuUOXusdEfFhSErRuDR9+aM4Q++47qysSkUJSEDqDq0XI02sIudhsNuqcXj9olwZMi/gew4A334Qrr4Q//oCaNc0Q1K2b1ZWJSCEpCJ3Bvc9YKUydd6kT7Zo5ll5q7ykiJSAlBfr0geHDISsLevaEjRvhqqusrkxEikBB6AyppbTP2Jni3GsJnSq19xSREjByJHz8MQQGwssvwxdfmLvHi4hP0ayxM5TWPmNncu9CrxYhEd8ycaI5HmjaNHPVaBHxSWoROkNKKQ+WhrwgtENjhES82/Hj5irRLpdcAuvWKQSJ+DgFoTPkjREq/a6x5FNZnDzdIiUiXubHH6FlSxg0yOwCcymF9cZExLMUhM7g7horxRahyNAgYsoHA+oeE/E6hmGO/+nQAf76C+rVM1uCRKTMUBA6gxVBCPK6x3ZqLSER73H0KNx8Mzz6KOTkQO/esGGDuV6QiJQZhe4Duu222wr9op999lmxirGaFdPnAeKiy7Fu13FttSHiLVatgjvugL17ISQEpk6Fe+5RV5hIGVToIFShQgVP1uEV8hZULN3JdHUqn545phYhEe+wf78ZgurXh48+ghYtrK5IRDyk0L/xZ82a5ck6LGcYRt46QqXcIlQn2rWWkIKQiGUMI6/F55//hNmz4bbbICLC0rJExLM0Rui0U1k5OA3zz6U9RihvUUUFIRFL/O9/5tifAwfyjg0cqBAk4gcK3SLUsmVLbIXsH9+wYUOxC7JKaqY5Pig4wE5IYOnmw7jTLUIpGQ6Op2VTsVxwqb6/iN/KzYUJE2DcOHA6YcwYmDHD6qpEpBQVOgj16tXLg2VY78ztNQob+EpKWHAAsRVCOZCSyY7kNForCIl43sGDcNddsGyZ+XjQIHNQtIj4lUIHobFjx3qyDstZNT7IpU5MOQ6kZLIrOY3WtStaUoOI31i2DO68Ew4dgvBwcwf5AQOsrkpELKAxQqe5usZKe3yQi8YJiZSSBQvg+uvNENSkCfz0k0KQiB8r1jzx3NxcXnnlFT766CN2795NdnZ2vuePHTtWIsWVplQL9hk7U10tqihSOq6/Hi67DDp2hFdfhbAwqysSEQsVq0Vo/PjxTJkyhT59+pCSkkJCQgK33XYbdrudcePGlXCJpSNv5/nSXUPIxTVgWosqinjAunXmYGiA8uXhhx/grbcUgkSkeEFo7ty5zJgxg0ceeYTAwED69u3L22+/zZgxY/jhhx9KusZS4V5V2qIWoTMXVTQMw5IaRMqcnBwYPRri42HKlLzjfrBArIgUTrGC0MGDB2natCkA5cuXJyUlBYAePXrw1VdflVx1pSjF4sHSNSuGY7dBenYuh09mWVKDSJmyZw906gSTJpmP9+61tBwR8U7FCkKXXHIJB04vPFavXj2+/fZbANatW0dISEjJVVeK8jZctaZrLDjQziUVwwENmBa5aF99ZW6LsWoVREbCxx9raryIFKhYQejWW29l2em1N+6//36eeeYZ6tevz4ABA/jXv/5VogWWFqunz0PeLvS7FIREiic729wtvkcPOHYM2rSBjRvhH/+wujIR8VLFav6Y5GpqBvr06UPt2rVZvXo19evXp2fPniVWXGnK23DV2iD0v21H1CIkUlybN8N//mP++cEH4YUXzN3jRUTOoUT6ga688kquvPLKkngpy1g9WBryWoQUhESKqXlzeP11qFIFyvhq+CJSMorVNTZx4kRmzpx51vGZM2fywgsvXHRRVrB6+jxoUUWRIsvKgkcegaSkvGPDhikEiUihFSsI/fe//6Vhw4ZnHW/cuDHTp0+/6KKsYPWCipC3qOJfx9LJdWoKvch5bd8O7dub0+L79AGHw+qKRMQHFXv6fGxs7FnHK1eu7J5N5kucToOTWae7xiwcLF09KozgADvZOU72n8iwrA4Rr/fxx9CqFaxfD5UqmWEoyLq/uyLiu4oVhGrWrMmqVavOOr5q1SqqV69+0UWVtpNZObjWMIywsGsswG6jVrQ5hX6XttoQOVtmJgwfDr17Q2qq2SKUlAQ33WR1ZSLio4r1W3/o0KE89NBDOBwOrr32WgCWLVvG448/ziOPPFKiBZYGV7dYSKCd0KAAS2uJiy7Hn4dPsTM5jY71K1tai4hXOXIEbrghbzzQ6NHw7LMQaN0/XkTE9xXr/yCPPfYYR48eZfjw4e4NV0NDQxk1ahSjR48u0QJLgzdMnXepE6NFFUUKVKkSxMRA5crw3nvQtavVFYlIGVCsIGSz2XjhhRd45pln2Lx5M2FhYdSvX993V5X2gqnzLnViygNaVFEEgPR0sNnMzVEDAmDuXHP/MB/sghcR71SsMUIuBw8e5NixY9SrV4+QkBCf3SzUG6bOu8SpRUjEtHkztG0LDz2Ud6xKFYUgESlRxQpCR48e5brrrqNBgwZ0797dPVNsyJAhPj1GyBtahOqebhHaczwDR67T4mpELDJnjrk9xm+/wRdfmOODREQ8oFhB6OGHHyYoKIjdu3cTHh7uPt6nTx8WL15cYsWVFqt3nj9T1cgQwoICyHUa7DmWbnU5IqUrLQ0GDTK/0tPhuuvMwdGVNXFARDyjWEHo22+/5YUXXuCSSy7Jd7x+/fr89ddfJVJYaUrNdI0Rsr5rzGazuVeY1hR68Su//QZXXGG2Btnt8Nxz8M03UK2a1ZWJSBlWrCCUlpaWryXI5dixYz45YNobdp4/k2vm2I4jCkLiJ7Kz4cYbzXFB1avD8uXw9NPmAGkREQ8qVhDq2LEj7777rvuxzWbD6XTy4osv0rlz5xIrrrS4B0t7wRghyNt8VS1C4jeCg2H6dDMMJSXBNddYXZGI+Ili9QW99NJLXHvttfz0009kZ2fz+OOP8/vvv3Ps2LECV5z2dq7p896wjhCYiyqCZo5JGffzz3D4MFx/vfn4ppuge3dzuryISCkpcouQw+HggQce4P/+7//o0KEDt9xyC2lpadx2221s3LiRevXqeaJOj8qbPu8dQcjdIpSswdJSBhmG2frTtq25Weru3XnPKQSJSCkrcotQUFAQv/zyCxUrVuSpp57yRE2lLm/6vPWDpSEvCO1PySDTkWv5th8iJSYlBYYNg48+Mh9ffz2UK2dtTSLi14o1Ruiuu+7inXfeKelaLONtg6UrlQsmIjQQw4C/jqpVSMqI9evNHeM/+sjcH+zll2HhQoiOtroyEfFjxWoCycnJYebMmSxdupTWrVtT7m//opsyZUqJFFda8qbPe0cQstls1I0px897U9iZnMZl1SKsLknk4rz2Gjz6qDk7rHZtmD/f7BoTEbFYsYLQb7/9RqtWrQDYtm1bvudsPtbHn5Pr5FTW6SDkBVtsuMSdEYREfN7vv5shqFcvmDkTKla0uiIREaCYQWjFihUlXYdlXCEIvKdFCM4cMK0gJD7KMPIGP7/yClx1FfTvrwHRIuJVLmrT1bLANXU+PDiAoADv+Xa4gpBahMTnGAZMmWJOhc/NNY+FhcGAAQpBIuJ1vKcvyCLeNnXexR2EtKii+JKjR819wr780nz82Wfwz39aWpKIyPl4TxOIRbxt6ryLa7+xIyezOHk6rIl4tdWroWVLMwSFhMCbb8I//mF1VSIi5+X3Qcibdp4/U2RoENHlggFNoRcv53TCCy/A1VfDnj1Qvz788APce6+6wkTE6/l9EPK2fcbOpHFC4hMeeACeeMIcD9Svn7leUIsWVlclIlIoCkIZ3jd13iVOQUh8wbBhUKkSvP02vP8+RGjdKxHxHd7327+U+UKLkKbQi1fJzYWffspbELFZM9i1SwFIRHySWoS8dIwQ5AWhHQpC4i0OHYJu3aBDB/jxx7zjCkEi4qMUhE5vr1HBm1uENIVevMHy5dC8OSxdCsHBsHev1RWJiFw0BSEvnT4PEBdtBqET6Q6Op2VbXI34rdxcGDsWunQxW4SaNDG7xm6/3erKREQumoKQly6oCBAWHEBshVBACyuKRfbvNwPQs8+aK0bffbfZJdaokdWViYiUCL8PQu51hLywawzyWoV2HlEQEgt89hkkJkL58jB3LsyYAeHhVlclIlJivCIITZs2jbi4OEJDQ2nbti1r164t1HXz5s3DZrPRq1evYr933vR5Lw1CGickVhoxAh591FwbqF8/q6sRESlxlgeh+fPnk5CQwNixY9mwYQPNmzena9euHD58+LzX7dq1i0cffZSOHTte1PvnTZ/3vjFCAHU1c0xKUWhyMgFDhsDJk+YBmw1eegkaNLC2MBERD7E8CE2ZMoWhQ4cyePBgLr/8cqZPn054eDgzZ8485zW5ubnceeedjB8/nrp16xb7vR25TtKzzd2xvb5FSEFIPMy2aBGdHn4Y+3vvwSOPWF2OiEipsLQZJDs7m/Xr1zN69Gj3MbvdTpcuXVizZs05r3v22WepUqUKQ4YMYeXKled9j6ysLLKystyPU1NTAXA4HBw7meE+Hhpg4HB43+amNaNCADMIZWdnYytDeze5vt/e+H33Kw4H9meeIXDKFAKB3JYtcSYkgO6LZfR3w3voXngPT90DS4NQcnIyubm5VK1aNd/xqlWrsmXLlgKv+f7773nnnXdISkoq1HtMnDiR8ePHn3V8xYoVpNnCgUBCAgy+/WZxUcsvFTlOsBFAWnYu8774mgrBVldU8pYsWWJ1CX4r7PBh2rz8MpW2bgVge48ebBo4EOfWrXD6mFhHfze8h+6F9dLTPbMBuXcOjDmHkydP0r9/f2bMmEFMTEyhrhk9ejQJCQnux6mpqdSsWZPOnTuzPzMQkn6kUvkwune/2lNlX7Sp21ay53gG9VpcSXxcJavLKTEOh4MlS5Zw/fXXExTknV2TZZnt++8JGDQI24kTGFFRZE+fzm+hobofXkB/N7yH7oX3OHr0qEde19IgFBMTQ0BAAIcOHcp3/NChQ1SrVu2s87dv386uXbvo2bOn+5jT6QQgMDCQrVu3Uq9evXzXhISEEBISctZrBQUFkXbSvLZCWJBX/4DXig5nz/EMDp10eHWdxRUU5N3f/zKrUSMICYG2bbHNm4e9Rg1YtEj3w4voXngP3Qvreer7b+lg6eDgYFq3bs2yZcvcx5xOJ8uWLaNdu3Znnd+wYUN+/fVXkpKS3F8333wznTt3JikpiZo1axbp/b196rxLWFAAAFk5TosrEZ935r+oqlUz1wj67juIi7OqIhERS1neNZaQkMDAgQNp06YN8fHxTJ06lbS0NAYPHgzAgAEDqFGjBhMnTiQ0NJQmTZrkuz4qKgrgrOOF4e1T511CAs0glK0gJBfjk09gyBB46y3o08c81rChtTWJiFjM8gTQp08fjhw5wpgxYzh48CAtWrRg8eLF7gHUu3fvxm73TMOVN+88f6aQQPPzZ+XkWlyJ+KTMTHM6/BtvmI/nzIHevc01gkRE/JzlQQhg5MiRjBw5ssDnEhMTz3vt7Nmzi/2+eS1CXh6Egk4HIYdahKSI/vjDDD2uWZZPPGHuG6YQJCICeEkQsop7jJCXB6HgAFeLkIKQFMGHH8KwYXDqFMTEwHvvQbduVlclIuJV/DsIuXee9+5vQ8jpwdLZuQpCUki//JK3N9jVV8MHH0CNGtbWJCLihbw7AXiYt+887+IeI+TQGCEppGbNzM1Sw8JgzBgI9Ou/6iIi5+TX/3f0vcHSahGS85g7Fzp2hFq1zMcvvqixQCIiF2D5pqtWSs10jRHy7jwYfDoIafq8FCgtDf71L7jrLujbN2+PMIUgEZEL8u4E4GG+0yKkBRXlHH7/3ZwVtmkT2O3Qtav5XxERKRT/DkKnB0tX8JUxQlpHSFwMA2bNgpEjISMDYmPNAdGdOlldmYiIT/HbIJSV4yTz9Lo83j5YOlhjhORMaWlw773w/vvm465d4d13oUoVa+sSEfFBftuGnna6Nchmg4gQ786D6hqTfOx2c3p8QABMnAiLFikEiYgUk3cnAA9yDZQuHxKI3e7dg0o1a0wwDPPLbjenxH/0ERw5Ah06WF2ZiIhP898WoWxzvE15L28NgjO32NAYIb+UkgJ33AETJuQdu+wyhSARkRLgt0HIcbp1xTX+xpu5ttjQ9Hk/tH49tG5ttgD9+99w4IDVFYmIlCnenwI8JNt5OggFeP+3wLXFhrrG/IhhwGuvwVVXwfbtULs2rFhhzg4TEZES4/39Qh6S7UMtQhoj5GdOnIAhQ+Czz8zHvXrBzJlQsaKVVYmIlEl+G4QcOQbga0FIY4TKvJwcsxVo82YICoLJk+H++7VKtIiIh3h/CvAQ107uvtA1pi02/EhgIDz4INStC6tXwwMPKASJiHiQ96cAD/GtrrG8MUKGYVhcjZS4Y8fMrTJchg0z1wlq08a6mkRE/IT3pwAPcZweLB3iC0EoKK9GV0uWlBGrV0OLFtCjhzk2CMwWoHLlrKxKRMRveH8K8BBfahE6s/tOA6bLCKcTXngBrr4a9uwxxwMdPmx1VSIifsdvB0tn554eLO0DY4TObLXSOKEy4MgRGDgQvv7afNy3L/z3vxARYW1dIiJ+yPtTgIe4FlR0jb/xZjabTRuvlhXffWd2hX39NYSGwowZMHeuQpCIiEX8t0XIh7rGwGwVys5xapsNXzdlCuzfDw0bmqtFN21qdUUiIn7Nf4OQ0/eC0Ek0WNrnvfOOOTX+2WehfHmrqxER8Xu+kQI8wJf2GoMzptA7FIR8yvLl8Mgj5pYZANHRZquQQpCIiFfw3xYhH1pQEbTNhs/JzTVbfZ57zgxBbdtC795WVyUiIn/jv0HIh7bYAM4YLK0xQl5v/364805ITDQfDxlirhMkIiJex2+DUN6sMd8IQiHaZsM3fPst3HWXOUW+XDlzWvydd1pdlYiInINvpAAP8L3B0nnbbIiXeukl6NbNDEHNm8OGDQpBIiJezjdSgAe4p8/7yhihIHWNeb2WLc3/3ncf/PADNGhgbT0iInJBfts15nNjhE4HNs0a8zKHD0OVKuafu3SBX3+Fxo2trUlERArNN1KAB7hnjflIEHK1CGkdIS/hcMBjj5mtPtu35x1XCBIR8Sm+kQI8wOe6xrSOkPf46y/o2BEmT4aUFPi//7O6IhERKSa/7Rpz+FqLkKbPe4fPP4fBg+HECahQAWbOhNtus7oqEREpJt9IAR7ga11jwZo+b63sbHjoIbj1VjMExcfDxo0KQSIiPs43UoAHZPvoOkKaPm+R11+HV181/5yQACtXQp061tYkIiIXzc+7xuwEBwRYXUqhaB0hi40cCUuWwPDh0LOn1dWIiEgJ8Y3mEA/I9rlNVzVGqFRlZpqbozoc5uPgYPj6a4UgEZEyxm9bhHxtZelgdY2Vnj/+gD59zDFAR47AxIlWVyQiIh7iGynAAxw+tqCixgiVknnzoFUrMwTFxMDVV1tdkYiIeJBvpAAPcOSeDkK+so5QkNYR8qiMDLjnHujbF06dMtcJSkqCG2+0ujIREfEg30gBHuRasdnbuQKbVpb2gG3boG1beOstsNng6adh+XKoUcPqykRExMP8doyQi++0CLn2GtNg6RLndMKOHeaeYXPnmnuGiYiIX1AQ8pUgpOnzJcvpBPvpe9+wIXz2GTRtCrGx1tYlIiKlyjdSgIcEBdiw221Wl1EoGixdgn7/HVq0gO++yzt2ww0KQSIifsivg5CvtAbBmVtsqGus2AwD3nkHrrgCfv0VHnnEPCYiIn7Ld5KAB/jK1HlQi9BFO3kS+veHu+82Z4jdcAN89ZU5OFpERPyW7yQBD/CtIKQxQsX288/Qpo05EDogACZMMFeJrlLF6spERMRifj1Y2peCkHafL6bNm82p8VlZ5nT4efOgQwerqxIRES/h30HIh8YIaa+xYmrYEG6+GdLSYM4cc7VoERGR0/w7CAX6xs7zcMY6QjlODMPAprEt57ZxI9SpA1FR5higOXMgJCRvuryIiMhpfv2bwZe6xlxjhAwjb3sQ+RvDgNdfhyuvNAdFu2aEhYUpBImISIH8ukUoxAe7xsDcZsOXQlypOHEChgwxF0YEyMmBzEwzBImIiJyDX/829aUwceZ4Jm2z8Tdr10LLlmYICgqCqVNhwQKFIBERuSDfSQIe4EtByG63ucOQptCfZhjwyivmLLBdu8xxQatWwYMPan0gEREpFN9JAh7gS7PGQIsqniUlBaZMAYcDbr8dNmwwV40WEREpJL8eI+RLLUJwut4srSXkFhUFH35oLpg4fLhagUREpMgUhHyI368l5HTC5MlQrRoMGGAe69BBCySKiEixKQj5kJAgP95m48gRGDjQ3BojPBw6d4aaNa2uSkREfJx/ByEfGyPkqtfvusZWroQ77oD9+yE01JwVdsklVlclIiJlgG8lgRIW4nMtQn7WNeZ0wr//DZ06mSHossvgxx9h6FCNBxIRkRLh1y1CPheEXGOEHH7QIpSbCzfdBN98Yz7u3x/eeAPKl7e2LhERKVN8KwmUMJ8bIxToR2OEAgKgTRtzPNCsWfDuuwpBIiJS4nwrCZQwXwtCrnrL7Bih3FxzULTLuHGQlASDBllUkIiIlHVekQSmTZtGXFwcoaGhtG3blrVr157z3BkzZtCxY0cqVqxIxYoV6dKly3nPPx9fGyxdpqfPHzgA118PN94IWVnmscBAqF/f2rpERKRMszwJzJ8/n4SEBMaOHcuGDRto3rw5Xbt25fDhwwWen5iYSN++fVmxYgVr1qyhZs2a3HDDDezbt6/I7x18uqvJV5TVlaVtS5ZA8+awYgVs2WIukCgiIlIKLA9CU6ZMYejQoQwePJjLL7+c6dOnEx4ezsyZMws8f+7cuQwfPpwWLVrQsGFD3n77bZxOJ8uWLSvye/ta11iZGyOUk0Oj998noEcPs0usWTNYvx7i462uTERE/ISls8ays7NZv349o0ePdh+z2+106dKFNWvWFOo10tPTcTgcVKpUqcDns7KyyHJ1tQCpqanuPwfgxOFwFLP60ufKbRlZDp+qu0B792Lv358Gq1YBkDt0KM7Jk80d4339s/ko18+Uz/9slQG6F95D98J7eOoeWBqEkpOTyc3NpWrVqvmOV61alS1bthTqNUaNGkX16tXp0qVLgc9PnDiR8ePHF/jcrz8nYd+7sWhFW2jfHjtgZ/O2P1mUtc3qci7Klc8+S9UNG3CEhZE0YgT7O3Qwu8bEckuWLLG6BDlN98J76F5YLz093SOv69PrCE2aNIl58+aRmJhIaGhogeeMHj2ahIQE9+PU1FRqnt6aoV3bNnRqULlUai0JW5b8QeKBndSoFUf37g2tLufiNGxI7rBh/K9vX64aMIAWQUFWV+T3HA4HS5Ys4frrrydI98NSuhfeQ/fCexw9etQjr2tpEIqJiSEgIIBDhw7lO37o0CGqVat23msnT57MpEmTWLp0Kc2aNTvneSEhIYSEhBT4XHhIsE/9YIcGm7XmGPhU3QDs3g3ffgt3320+vuwyHEuXkrZoEUFBQb73ecow3Q/voXvhPXQvrOep77+lo4WDg4Np3bp1voHOroHP7dq1O+d1L774Is899xyLFy+mTZs2xX9/XxssHeSjK0svXAgtWsCwYWYYEhER8RKWd40lJCQwcOBA2rRpQ3x8PFOnTiUtLY3BgwcDMGDAAGrUqMHEiRMBeOGFFxgzZgwffPABcXFxHDx4EIDy5ctTvogrD2sdIQ/LzoZRo8xNUgGuuELrAomIiFexPAj16dOHI0eOMGbMGA4ePEiLFi1YvHixewD17t27sdvzAsubb75JdnY2//jHP/K9ztixYxk3blyR3tvnWoR8afr8zp3Qpw+sW2c+fvhhmDQJgoOtrUtEROQMlgchgJEjRzJy5MgCn0tMTMz3eNeuXSX2vr4WhHxmi43PPze3xUhJgYoVYfZsuPlmi4sSERE5m1cEIauoa8xDUlPNENSuHcybB7VqWV2RiIhIgfw6CIX4WIuQV2+xkZtr7hgPMGAAhIbCrbeCZlmIiIgX860kUMJ8rWssJMgMGl7XNTZvHjRtCsnJecd691YIEhERr+dbSaCE+VoQcnXleU2LUEYG3HMP9O0LmzfDlClWVyQiIlIkft015nNjhIK8aIzQli1mq8+vv4LNBk8+CUWctSciImI1vw1CAXYbgb4WhAK9ZEHF996D++6DtDSoUgXefx+uv97amkRERIrBb4NQUIDN6hKKzBWEsnMtDEL//S/ce6/5586dYe5ciI21rh4REZGL4FtNIiUo2CeD0OkFFa1sEbrjDrj0UrMbbMkShSAREfFpftsi5Gvjg8CidYQMA5Yvh2uvNccCVagAv/wCYWGlV4OIiIiH+F4aKCFBPjZjDPJahJwG5JRG99ipUzBwIHTpAtOn5x1XCBIRkTJCLUI+5Mzp/lk5Ts8O9v7lF3NW2NatYLebA6NFRETKGN9LAyXE19YQgrODkEcYhjkgOj7eDEE1akBiIjz6qGfeT0RExEJ+2yIU5IMtQgF2G0EBNhy5hmfGCaWmwrBhMH+++fjGG+HddyEmpuTfS0RExAv4XhooIb7YNQZ544Q8ss3Gb7/Bxx+be4a9+CJ8+aVCkIiIlGl+2yLki11jcLruLA91jV11Fbz+OrRoYe4cLyIiUsb5ZhooAb7YNQYlvLr0iRPQv7+5T5jLffcpBImIiN/w4xYh31tQEUpwLaF166BPH9i5EzZtgp9+MtcJEhER8SO+2SxSAnx1jJCrS6/YY4QMA6ZOhfbtzRAUF2euEaQQJCIifshvW4R8ca8xOGObjeIEoWPHYPBgWLjQfHzbbfDOOxAVVXIFioiI+BC/DUK+Oli62F1jO3dCp06wezcEB8OUKTB8uFqCRETEr/lvEPLRrrGQIFcQKmKLUM2aUKsWBAXBRx9Bq1YeqE5ERMS3+G0Q8sW9xiAvwBUqCB09ChERZgtQYKC5RlB4OERGerhKERER3+CbaaAE+GyLUGHHCK1cCc2bw6hReceqVVMIEhEROYNvpoES4LPrCLm6xhznGCPkdMKECdC5M+zbB4sXa8NUERGRc/DNNFACfDUIuVqysnMLaBE6fBi6dYOnnoLcXLjrLnO9oHLlSrlKERER3+C3Y4RCgnxztlRei9DfgtCKFdCvHxw8CGFhMG0aDBqkWWEiIiLn4bdBqEyNEUpNhdtvh+PH4fLLzVlhjRtbVKGIiIjv8Nsg5KuzxkIKWlk6MhL++1/4+mt47TV1hYmIiBSS3wahYLtvBiHXQpCXrF8F5Q7CtdeaT/zzn+aXiIiIFJr/BiEfbREKtRk88t17DPrhI6hcGZKSIDbW6rJERER8kt8GoajwIKtLKLp9++j1SH+q/bzOfNyrl/YJExERuQi+2SxSAtrUrmh1CUXz9dfQogXVfl6HUb48Oe/PNccFhYVZXZmIiIjP8tsg5DOcTnN16O7dITkZWrbEtmEDgXf2s7oyERERn6cg5O3sdnNtIIARI2D1aqhf39qaREREygi/HSPk9XJyzI1SwVwc8Z//hB49rK1JRESkjFGLkLfJzoaEBLjtNjAM81j58gpBIiIiHqAWIW+ycyf06WPuDwaQmGhunioiIiIeoRYhb/HZZ9CypRmCoqLg888VgkRERDxMQchqWVlw//3mXmEpKXDlleYiibfcYnVlIiIiZZ6CkNXuvBNef93882OPwXffQe3a1tYkIiLiJxSErDZqlLlFxpdfwosvQpAPrngtIiLiozRYurRlZMDatXDNNebjK66AHTsgNNTaukRERPyQWoRK09at5higrl3NcUAuCkEiIiKWUBAqLXPnQuvW8MsvEBkJJ05YXZGIiIjfUxDytPR0uPtuuOsuSEuDTp3M1qBOnSwuTERERBSEPGnTJoiPh3feAZsNxo6FpUuhenWrKxMRERE0WNqzvvgCfv8dqlUzu8auvdbqikREROQMCkKe9PjjZnfY/fdD1apWVyMiIiJ/o66xkvTrr+Yu8RkZ5uOAAHj+eYUgERERL6UgVBIMA2bMMMcDffIJjBtndUUiIiJSCOoau1ipqXDPPTBvnvm4Wzd49FFraxIREZFCUYvQxdi40VwbaN48sxvshRfgq6+gcmWrKxMREZFCUItQcS1YAHfcAdnZULOmGYauusrqqkRERKQIFISKq00bKF8e2reHWbMgOtrqikRERKSIFISKYt8+qFHD/HPNmubmqXXrmoslioiIiM/RGKHCMAx49VUz9CxcmHe8Xj2FIBERER+mIHQhx47BrbfCQw+Z44HODEIiIiLi0xSEzueHH6BlS3OrjOBgeO01c70gERERKRMUhAridMLkydCxI+zebXaBrV4NI0eqK0xERKQMURAqyHffwWOPQU4O9O4NGzaY6wWJiIhImaJZYwXp1AkefBAaNjRXjVYrkIiISJmkIARmV9irr0LfvlCtmnls6lRLSxIRERHP84qusWnTphEXF0doaCht27Zl7dq15z3/448/pmHDhoSGhtK0aVMWLVpU/Dc/fBhuvBESEuDOO81QJCIiIn7B8iA0f/58EhISGDt2LBs2bKB58+Z07dqVw4cPF3j+6tWr6du3L0OGDGHjxo306tWLXr168dtvvxX9zRMToUUL+PZbCAszg5C6wURERPyG5UFoypQpDB06lMGDB3P55Zczffp0wsPDmTlzZoHnv/rqq3Tr1o3HHnuMRo0a8dxzz9GqVStef/31Ir2v/aWX4Lrr4MABaNTIXCX6X/9SEBIREfEjlgah7Oxs1q9fT5cuXdzH7HY7Xbp0Yc2aNQVes2bNmnznA3Tt2vWc559LwAsvmN1ggwfDunXQpEnRP4CIiIj4NEsHSycnJ5Obm0vVqlXzHa9atSpbtmwp8JqDBw8WeP7BgwcLPD8rK4usrCz345SUFPO/oaHkvPwyRp8+kJlpfkmpcjgcpKenc/ToUYKCgqwux+/pfngP3QvvoXvhPY4dOwaAYRgl+rplftbYxIkTGT9+/FnHa2VmwogR5peIiIj4hKNHj1KhQoUSez1Lg1BMTAwBAQEcOnQo3/FDhw5RzTWN/W+qVatWpPNHjx5NQkKC+/GJEyeoXbs2u3fvLtFvpBRdamoqNWvWZM+ePURGRlpdjt/T/fAeuhfeQ/fCe6SkpFCrVi0qVapUoq9raRAKDg6mdevWLFu2jF69egHgdDpZtmwZI0eOLPCadu3asWzZMh566CH3sSVLltCuXbsCzw8JCSEkJOSs4xUqVNAPtZeIjIzUvfAiuh/eQ/fCe+heeA+7vWSHN1veNZaQkMDAgQNp06YN8fHxTJ06lbS0NAYPHgzAgAEDqFGjBhMnTgTgwQcf5JprruHll1/mpptuYt68efz000+89dZbVn4MERER8UGWB6E+ffpw5MgRxowZw8GDB2nRogWLFy92D4jevXt3vvR31VVX8cEHH/D000/z5JNPUr9+fT7//HOaaNaXiIiIFJHlQQhg5MiR5+wKS0xMPOvYP//5T/75z38W671CQkIYO3Zsgd1lUrp0L7yL7of30L3wHroX3sNT98JmlPQ8NBEREREfYfnK0iIiIiJWURASERERv6UgJCIiIn5LQUhERET8VpkMQtOmTSMuLo7Q0FDatm3L2rVrz3v+xx9/TMOGDQkNDaVp06YsWrSolCot+4pyL2bMmEHHjh2pWLEiFStWpEuXLhe8d1I0Rf274TJv3jxsNpt74VO5eEW9FydOnGDEiBHExsYSEhJCgwYN9P+qElLUezF16lQuu+wywsLCqFmzJg8//DCZ2q/yon333Xf07NmT6tWrY7PZ+Pzzzy94TWJiIq1atSIkJIRLL72U2bNnF/2NjTJm3rx5RnBwsDFz5kzj999/N4YOHWpERUUZhw4dKvD8VatWGQEBAcaLL75obNq0yXj66aeNoKAg49dffy3lysueot6Lfv36GdOmTTM2btxobN682Rg0aJBRoUIFY+/evaVcedlU1PvhsnPnTqNGjRpGx44djVtuuaV0ii3jinovsrKyjDZt2hjdu3c3vv/+e2Pnzp1GYmKikZSUVMqVlz1FvRdz5841QkJCjLlz5xo7d+40vvnmGyM2NtZ4+OGHS7nysmfRokXGU089ZXz22WcGYCxYsOC85+/YscMIDw83EhISjE2bNhmvvfaaERAQYCxevLhI71vmglB8fLwxYsQI9+Pc3FyjevXqxsSJEws8v3fv3sZNN92U71jbtm2Ne+65x6N1+oOi3ou/y8nJMSIiIow5c+Z4qkS/Upz7kZOTY1x11VXG22+/bQwcOFBBqIQU9V68+eabRt26dY3s7OzSKtFvFPVejBgxwrj22mvzHUtISDDat2/v0Tr9TWGC0OOPP240btw437E+ffoYXbt2LdJ7lamusezsbNavX0+XLl3cx+x2O126dGHNmjUFXrNmzZp85wN07dr1nOdL4RTnXvxdeno6DoejxDfY80fFvR/PPvssVapUYciQIaVRpl8ozr1YuHAh7dq1Y8SIEVStWpUmTZowYcIEcnNzS6vsMqk49+Kqq65i/fr17u6zHTt2sGjRIrp3714qNUuekvr97RUrS5eU5ORkcnNz3dtzuFStWpUtW7YUeM3BgwcLPP/gwYMeq9MfFOde/N2oUaOoXr36WT/oUnTFuR/ff/8977zzDklJSaVQof8ozr3YsWMHy5cv584772TRokX8+eefDB8+HIfDwdixY0uj7DKpOPeiX79+JCcn06FDBwzDICcnh3vvvZcnn3yyNEqWM5zr93dqaioZGRmEhYUV6nXKVIuQlB2TJk1i3rx5LFiwgNDQUKvL8TsnT56kf//+zJgxg5iYGKvL8XtOp5MqVarw1ltv0bp1a/r06cNTTz3F9OnTrS7N7yQmJjJhwgTeeOMNNmzYwGeffcZXX33Fc889Z3VpUkxlqkUoJiaGgIAADh06lO/4oUOHqFatWoHXVKtWrUjnS+EU5164TJ48mUmTJrF06VKaNWvmyTL9RlHvx/bt29m1axc9e/Z0H3M6nQAEBgaydetW6tWr59miy6ji/N2IjY0lKCiIgIAA97FGjRpx8OBBsrOzCQ4O9mjNZVVx7sUzzzxD//79ufvuuwFo2rQpaWlpDBs2jKeeeirfJuHiWef6/R0ZGVno1iAoYy1CwcHBtG7dmmXLlrmPOZ1Oli1bRrt27Qq8pl27dvnOB1iyZMk5z5fCKc69AHjxxRd57rnnWLx4MW3atCmNUv1CUe9Hw4YN+fXXX0lKSnJ/3XzzzXTu3JmkpCRq1qxZmuWXKcX5u9G+fXv+/PNPdxgF2LZtG7GxsQpBF6E49yI9Pf2ssOMKqIa27ixVJfb7u2jjuL3fvHnzjJCQEGP27NnGpk2bjGHDhhlRUVHGwYMHDcMwjP79+xtPPPGE+/xVq1YZgYGBxuTJk43NmzcbY8eO1fT5ElLUezFp0iQjODjY+OSTT4wDBw64v06ePGnVRyhTino//k6zxkpOUe/F7t27jYiICGPkyJHG1q1bjS+//NKoUqWK8fzzz1v1EcqMot6LsWPHGhEREcaHH35o7Nixw/j222+NevXqGb1797bqI5QZJ0+eNDZu3Ghs3LjRAIwpU6YYGzduNP766y/DMAzjiSeeMPr37+8+3zV9/rHHHjM2b95sTJs2TdPnXV577TWjVq1aRnBwsBEfH2/88MMP7ueuueYaY+DAgfnO/+ijj4wGDRoYwcHBRuPGjY2vvvqqlCsuu4pyL2rXrm0AZ32NHTu29Asvo4r6d+NMCkIlq6j3YvXq1Ubbtm2NkJAQo27dusa///1vIycnp5SrLpuKci8cDocxbtw4o169ekZoaKhRs2ZNY/jw4cbx48dLv/AyZsWKFQX+DnB9/wcOHGhcc801Z13TokULIzg42Khbt64xa9asIr+vzTDUliciIiL+qUyNERIREREpCgUhERER8VsKQiIiIuK3FIRERETEbykIiYiIiN9SEBIRERG/pSAkIiIifktBSERKnWEYDBs2jEqVKmGz2S64w/2uXbvynZeYmIjNZuPEiRMer1VEyjYFIREpdYsXL2b27Nl8+eWXHDhwgCZNmlhd0jnZbDY+//xzq8sQEQ8pU7vPi4hv2L59O7GxsVx11VVWlyIifk4tQiJSqgYNGsT999/P7t27sdlsxMXFsXjxYjp06EBUVBTR0dH06NGD7du3X/R7/e9//yM+Pp6QkBBiY2N54oknyMnJcT8fFxfH1KlT813TokULxo0b534e4NZbb3XXKiJli4KQiJSqV199lWeffZZLLrmEAwcOsG7dOtLS0khISOCnn35i2bJl2O12br31VpxOZ7HfZ9++fXTv3p0rrriCn3/+mTfffJN33nmH559/vtCvsW7dOgBmzZrlrlVEyhZ1jYlIqapQoQIREREEBARQrVo1AG6//fZ858ycOZPKlSuzadOmYo8feuONN6hZsyavv/46NpuNhg0bsn//fkaNGsWYMWOw2y/878DKlSsDEBUV5a5VRMoWtQiJiOX++OMP+vbtS926dYmMjHR3Qe3evbtQ15cvX979de+99wKwefNm2rVrh81mc5/Xvn17Tp06xd69e0v8M4iIb1KLkIhYrmfPntSuXZsZM2ZQvXp1nE4nTZo0ITs7u1DXnzn9PjIystDva7fbMQwj3zGHw1Ho60XE9ykIiYiljh49ytatW5kxYwYdO3YE4Pvvvy/Sa1x66aVnHWvUqBGffvophmG4W4VWrVpFREQEl1xyCWB2fR04cMB9TWpqKjt37sz3OkFBQeTm5hapHhHxHeoaExFLVaxYkejoaN566y3+/PNPli9fTkJCwkW/7vDhw9mzZw/3338/W7Zs4YsvvmDs2LEkJCS4xwdde+21vPfee6xcuZJff/2VgQMHEhAQkO914uLiWLZsGQcPHuT48eMXXZeIeBcFIRGxlN1uZ968eaxfv54mTZrw8MMP89JLL13069aoUYNFixaxdu1amjdvzr333suQIUN4+umn3eeMHj2aa665hh49enDTTTfRq1cv6tWrl+91Xn75ZZYsWULNmjVp2bLlRdclIt7FZvy9g1xERETET6hFSERERPyWgpCIiIj4LQUhERER8VsKQiIiIuK3FIRERETEbykIiYiIiN9SEBIRERG/pSAkIiIifktBSERERPyWgpCIiIj4LQUhERER8VsKQiIiIuK3/h/jXnzPPeHYfQAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "ROC curve of knn classifier using 12 most \"important\" features\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "print('ROC curve of knn classifier using all available features')\n", "knn = Knn_Classifier(X_unknown, X_labeled, y_labeled)\n", "knn.set_k(5)\n", "plot_measures(knn, y_unknown, all=False, plot_type='ROC')\n", "\n", "print(f'ROC curve of knn classifier using {n_dim} most \"important\" features')\n", "pca = PCA(n_components=n_dim)\n", "pca.fit(X_labeled)\n", "knn = Knn_Classifier(pca.transform(X_unknown), pca.transform(X_labeled), y_labeled)\n", "knn.set_k(5)\n", "plot_measures(knn, y_unknown, all=False, plot_type='ROC')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Submitting your solution\n", "\n", "As a last step, the notebook should be uploaded to Ilias such that we can grade it." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.19" } }, "nbformat": 4, "nbformat_minor": 4 }